github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-07 08:30:05 -05:00
Code Issues 14.5k Packages Projects Releases 120 Wiki Activity

[GH-ISSUE #14438] Crash using qwen3-code-next:lastest #9375

New Issue
Open
opened 2026-04-12 22:16:41 -05:00 by GiteaMirror · 6 comments
GiteaMirror commented 2026-04-12 22:16:41 -05:00
Owner
Copy Link

Originally created by @akoww on GitHub (Feb 26, 2026).
Original GitHub issue: https://github.com/ollama/ollama/issues/14438

What is the issue?

I downloaded the qwen3-coder-next:latest model and try to run it. loading into ram works but crashes on first request.
I used the official docker image 0.17.1 with Vulkan backend (CPU backend did also not work)

on the other hand: qwen3-next:latest works perfectly fine.

Relevant log output

Feb 26 11:27:35 akoww-a9mega ollama[349347]: [GIN] 2026/02/26 - 10:27:35 | 200 |      30.938µs |       10.88.0.1 | HEAD     "/"
Feb 26 11:27:35 akoww-a9mega ollama[349347]: [GIN] 2026/02/26 - 10:27:35 | 200 |   62.652505ms |       10.88.0.1 | POST     "/api/show"
Feb 26 11:27:35 akoww-a9mega ollama[349347]: [GIN] 2026/02/26 - 10:27:35 | 200 |    65.47915ms |       10.88.0.1 | POST     "/api/show"
Feb 26 11:27:35 akoww-a9mega ollama[349347]: time=2026-02-26T10:27:35.971Z level=INFO source=server.go:431 msg="starting runner" cmd="/usr/bin/ollama runner --ollama-engine --port 45095"
Feb 26 11:27:36 akoww-a9mega ollama[349347]: time=2026-02-26T10:27:36.014Z level=WARN source=cpu_linux.go:130 msg="failed to parse CPU allowed micro secs" error="strconv.ParseInt: parsing \"max\": invalid syntax"
Feb 26 11:27:36 akoww-a9mega ollama[349347]: time=2026-02-26T10:27:36.061Z level=INFO source=server.go:247 msg="enabling flash attention"
Feb 26 11:27:36 akoww-a9mega ollama[349347]: time=2026-02-26T10:27:36.061Z level=INFO source=server.go:431 msg="starting runner" cmd="/usr/bin/ollama runner --ollama-engine --model /root/.ollama/models/blobs/sha256-30e51a7cb1cf1333b9e298b90b4c7790fe2572d8736b002482a0ac96328a2ffb --port 34865"
Feb 26 11:27:36 akoww-a9mega ollama[349347]: time=2026-02-26T10:27:36.062Z level=INFO source=sched.go:491 msg="system memory" total="124.0 GiB" free="119.4 GiB" free_swap="0 B"
Feb 26 11:27:36 akoww-a9mega ollama[349347]: time=2026-02-26T10:27:36.062Z level=INFO source=sched.go:498 msg="gpu memory" id=00000000-c700-0000-0000-000000000000 library=Vulkan available="123.9 GiB" free="124.3 GiB" minimum="457.0 MiB" overhead="0 B"
Feb 26 11:27:36 akoww-a9mega ollama[349347]: time=2026-02-26T10:27:36.062Z level=INFO source=server.go:757 msg="loading model" "model layers"=49 requested=-1
Feb 26 11:27:36 akoww-a9mega ollama[349347]: time=2026-02-26T10:27:36.068Z level=INFO source=runner.go:1411 msg="starting ollama engine"
Feb 26 11:27:36 akoww-a9mega ollama[349347]: time=2026-02-26T10:27:36.068Z level=INFO source=runner.go:1446 msg="Server listening on 127.0.0.1:34865"
Feb 26 11:27:36 akoww-a9mega ollama[349347]: time=2026-02-26T10:27:36.073Z level=INFO source=runner.go:1284 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Enabled KvSize:262144 KvCacheType: NumThreads:16 GPULayers:49[ID:00000000-c700-0000-0000-000000000000 Layers:49(0..48)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
Feb 26 11:27:36 akoww-a9mega ollama[349347]: time=2026-02-26T10:27:36.089Z level=INFO source=ggml.go:136 msg="" architecture=qwen3next file_type=Q4_K_M name="" description="" num_tensors=843 num_key_values=38
Feb 26 11:27:36 akoww-a9mega ollama[349347]: load_backend: loaded CPU backend from /usr/lib/ollama/libggml-cpu-icelake.so
Feb 26 11:27:36 akoww-a9mega ollama[349347]: ggml_vulkan: Found 1 Vulkan devices:
Feb 26 11:27:36 akoww-a9mega ollama[349347]: ggml_vulkan: 0 = Radeon 8060S Graphics (RADV GFX1151) (radv) | uma: 1 | fp16: 1 | bf16: 0 | warp size: 64 | shared memory: 65536 | int dot: 1 | matrix cores: KHR_coopmat
Feb 26 11:27:36 akoww-a9mega ollama[349347]: load_backend: loaded Vulkan backend from /usr/lib/ollama/vulkan/libggml-vulkan.so
Feb 26 11:27:36 akoww-a9mega ollama[349347]: time=2026-02-26T10:27:36.112Z level=INFO source=ggml.go:104 msg=system CPU.0.SSE3=1 CPU.0.SSSE3=1 CPU.0.AVX=1 CPU.0.AVX2=1 CPU.0.F16C=1 CPU.0.FMA=1 CPU.0.BMI2=1 CPU.0.AVX512=1 CPU.0.AVX512_VBMI=1 CPU.0.AVX512_VNNI=1 CPU.0.LLAMAFILE=1 CPU.1.LLAMAFILE=1 compiler=cgo(gcc)
Feb 26 11:27:36 akoww-a9mega ollama[349347]: ggml_backend_vk_get_device_memory called: uuid 00000000-c700-0000-0000-000000000000
Feb 26 11:27:36 akoww-a9mega ollama[349347]: ggml_backend_vk_get_device_memory called: luid 0x0000000000000000
Feb 26 11:27:36 akoww-a9mega ollama[349347]: time=2026-02-26T10:27:36.272Z level=INFO source=runner.go:1284 msg=load request="{Operation:alloc LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Enabled KvSize:262144 KvCacheType: NumThreads:16 GPULayers:49[ID:00000000-c700-0000-0000-000000000000 Layers:49(0..48)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
Feb 26 11:27:36 akoww-a9mega ollama[349347]: ggml_backend_vk_get_device_memory called: uuid 00000000-c700-0000-0000-000000000000
Feb 26 11:27:36 akoww-a9mega ollama[349347]: ggml_backend_vk_get_device_memory called: luid 0x0000000000000000
Feb 26 11:27:39 akoww-a9mega ollama[349347]: time=2026-02-26T10:27:39.157Z level=INFO source=runner.go:1284 msg=load request="{Operation:commit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Enabled KvSize:262144 KvCacheType: NumThreads:16 GPULayers:49[ID:00000000-c700-0000-0000-000000000000 Layers:49(0..48)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
Feb 26 11:27:39 akoww-a9mega ollama[349347]: time=2026-02-26T10:27:39.157Z level=INFO source=ggml.go:482 msg="offloading 48 repeating layers to GPU"
Feb 26 11:27:39 akoww-a9mega ollama[349347]: time=2026-02-26T10:27:39.157Z level=INFO source=ggml.go:489 msg="offloading output layer to GPU"
Feb 26 11:27:39 akoww-a9mega ollama[349347]: time=2026-02-26T10:27:39.157Z level=INFO source=ggml.go:494 msg="offloaded 49/49 layers to GPU"
Feb 26 11:27:39 akoww-a9mega ollama[349347]: time=2026-02-26T10:27:39.157Z level=INFO source=device.go:240 msg="model weights" device=Vulkan0 size="48.0 GiB"
Feb 26 11:27:39 akoww-a9mega ollama[349347]: time=2026-02-26T10:27:39.157Z level=INFO source=device.go:245 msg="model weights" device=CPU size="166.9 MiB"
Feb 26 11:27:39 akoww-a9mega ollama[349347]: time=2026-02-26T10:27:39.157Z level=INFO source=device.go:251 msg="kv cache" device=Vulkan0 size="7.8 GiB"
Feb 26 11:27:39 akoww-a9mega ollama[349347]: time=2026-02-26T10:27:39.157Z level=INFO source=device.go:262 msg="compute graph" device=Vulkan0 size="808.1 MiB"
Feb 26 11:27:39 akoww-a9mega ollama[349347]: time=2026-02-26T10:27:39.157Z level=INFO source=device.go:267 msg="compute graph" device=CPU size="9.0 MiB"
Feb 26 11:27:39 akoww-a9mega ollama[349347]: time=2026-02-26T10:27:39.157Z level=INFO source=device.go:272 msg="total memory" size="56.8 GiB"
Feb 26 11:27:39 akoww-a9mega ollama[349347]: time=2026-02-26T10:27:39.157Z level=INFO source=sched.go:566 msg="loaded runners" count=1
Feb 26 11:27:39 akoww-a9mega ollama[349347]: time=2026-02-26T10:27:39.157Z level=INFO source=server.go:1350 msg="waiting for llama runner to start responding"
Feb 26 11:27:39 akoww-a9mega ollama[349347]: time=2026-02-26T10:27:39.157Z level=INFO source=server.go:1384 msg="waiting for server to become available" status="llm server loading model"
Feb 26 11:27:51 akoww-a9mega ollama[349347]: time=2026-02-26T10:27:51.942Z level=INFO source=server.go:1388 msg="llama runner started in 15.88 seconds"
Feb 26 11:27:51 akoww-a9mega ollama[349347]: [GIN] 2026/02/26 - 10:27:51 | 200 | 16.042613048s |       10.88.0.1 | POST     "/api/generate"
Feb 26 12:19:03 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:03.591Z level=INFO source=server.go:431 msg="starting runner" cmd="/usr/bin/ollama runner --ollama-engine --port 37115"
Feb 26 12:19:03 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:03.639Z level=WARN source=cpu_linux.go:130 msg="failed to parse CPU allowed micro secs" error="strconv.ParseInt: parsing \"max\": invalid syntax"
Feb 26 12:19:03 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:03.685Z level=INFO source=server.go:247 msg="enabling flash attention"
Feb 26 12:19:03 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:03.685Z level=INFO source=server.go:431 msg="starting runner" cmd="/usr/bin/ollama runner --ollama-engine --model /root/.ollama/models/blobs/sha256-30e51a7cb1cf1333b9e298b90b4c7790fe2572d8736b002482a0ac96328a2ffb --port 38535"
Feb 26 12:19:03 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:03.686Z level=INFO source=sched.go:491 msg="system memory" total="124.0 GiB" free="119.5 GiB" free_swap="0 B"
Feb 26 12:19:03 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:03.686Z level=INFO source=sched.go:498 msg="gpu memory" id=00000000-c700-0000-0000-000000000000 library=Vulkan available="123.9 GiB" free="124.3 GiB" minimum="457.0 MiB" overhead="0 B"
Feb 26 12:19:03 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:03.686Z level=INFO source=server.go:757 msg="loading model" "model layers"=49 requested=-1
Feb 26 12:19:03 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:03.694Z level=INFO source=runner.go:1411 msg="starting ollama engine"
Feb 26 12:19:03 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:03.694Z level=INFO source=runner.go:1446 msg="Server listening on 127.0.0.1:38535"
Feb 26 12:19:03 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:03.698Z level=INFO source=runner.go:1284 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Enabled KvSize:262144 KvCacheType: NumThreads:16 GPULayers:49[ID:00000000-c700-0000-0000-000000000000 Layers:49(0..48)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
Feb 26 12:19:03 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:03.716Z level=INFO source=ggml.go:136 msg="" architecture=qwen3next file_type=Q4_K_M name="" description="" num_tensors=843 num_key_values=38
Feb 26 12:19:03 akoww-a9mega ollama[349347]: load_backend: loaded CPU backend from /usr/lib/ollama/libggml-cpu-icelake.so
Feb 26 12:19:03 akoww-a9mega ollama[349347]: ggml_vulkan: Found 1 Vulkan devices:
Feb 26 12:19:03 akoww-a9mega ollama[349347]: ggml_vulkan: 0 = Radeon 8060S Graphics (RADV GFX1151) (radv) | uma: 1 | fp16: 1 | bf16: 0 | warp size: 64 | shared memory: 65536 | int dot: 1 | matrix cores: KHR_coopmat
Feb 26 12:19:03 akoww-a9mega ollama[349347]: load_backend: loaded Vulkan backend from /usr/lib/ollama/vulkan/libggml-vulkan.so
Feb 26 12:19:03 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:03.742Z level=INFO source=ggml.go:104 msg=system CPU.0.SSE3=1 CPU.0.SSSE3=1 CPU.0.AVX=1 CPU.0.AVX2=1 CPU.0.F16C=1 CPU.0.FMA=1 CPU.0.BMI2=1 CPU.0.AVX512=1 CPU.0.AVX512_VBMI=1 CPU.0.AVX512_VNNI=1 CPU.0.LLAMAFILE=1 CPU.1.LLAMAFILE=1 compiler=cgo(gcc)
Feb 26 12:19:03 akoww-a9mega ollama[349347]: ggml_backend_vk_get_device_memory called: uuid 00000000-c700-0000-0000-000000000000
Feb 26 12:19:03 akoww-a9mega ollama[349347]: ggml_backend_vk_get_device_memory called: luid 0x0000000000000000
Feb 26 12:19:03 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:03.903Z level=INFO source=runner.go:1284 msg=load request="{Operation:alloc LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Enabled KvSize:262144 KvCacheType: NumThreads:16 GPULayers:49[ID:00000000-c700-0000-0000-000000000000 Layers:49(0..48)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
Feb 26 12:19:03 akoww-a9mega ollama[349347]: ggml_backend_vk_get_device_memory called: uuid 00000000-c700-0000-0000-000000000000
Feb 26 12:19:03 akoww-a9mega ollama[349347]: ggml_backend_vk_get_device_memory called: luid 0x0000000000000000
Feb 26 12:19:06 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:06.762Z level=INFO source=runner.go:1284 msg=load request="{Operation:commit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Enabled KvSize:262144 KvCacheType: NumThreads:16 GPULayers:49[ID:00000000-c700-0000-0000-000000000000 Layers:49(0..48)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
Feb 26 12:19:06 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:06.762Z level=INFO source=device.go:240 msg="model weights" device=Vulkan0 size="48.0 GiB"
Feb 26 12:19:06 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:06.762Z level=INFO source=device.go:245 msg="model weights" device=CPU size="166.9 MiB"
Feb 26 12:19:06 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:06.762Z level=INFO source=ggml.go:482 msg="offloading 48 repeating layers to GPU"
Feb 26 12:19:06 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:06.762Z level=INFO source=ggml.go:489 msg="offloading output layer to GPU"
Feb 26 12:19:06 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:06.762Z level=INFO source=ggml.go:494 msg="offloaded 49/49 layers to GPU"
Feb 26 12:19:06 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:06.762Z level=INFO source=device.go:251 msg="kv cache" device=Vulkan0 size="7.8 GiB"
Feb 26 12:19:06 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:06.762Z level=INFO source=device.go:262 msg="compute graph" device=Vulkan0 size="808.1 MiB"
Feb 26 12:19:06 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:06.762Z level=INFO source=device.go:267 msg="compute graph" device=CPU size="9.0 MiB"
Feb 26 12:19:06 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:06.762Z level=INFO source=device.go:272 msg="total memory" size="56.8 GiB"
Feb 26 12:19:06 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:06.762Z level=INFO source=sched.go:566 msg="loaded runners" count=1
Feb 26 12:19:06 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:06.762Z level=INFO source=server.go:1350 msg="waiting for llama runner to start responding"
Feb 26 12:19:06 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:06.763Z level=INFO source=server.go:1384 msg="waiting for server to become available" status="llm server loading model"
Feb 26 12:19:09 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:09.269Z level=WARN source=server.go:1357 msg="client connection closed before server finished loading, aborting load"
Feb 26 12:19:09 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:09.269Z level=ERROR source=sched.go:572 msg="error loading llama server" error="timed out waiting for llama runner to start: context canceled"
Feb 26 12:19:09 akoww-a9mega ollama[349347]: [GIN] 2026/02/26 - 11:19:09 | 499 |  5.737795975s |       10.88.0.1 | POST     "/api/chat"
Feb 26 12:19:36 akoww-a9mega ollama[349347]: [GIN] 2026/02/26 - 11:19:36 | 200 |      49.803µs |       10.88.0.1 | HEAD     "/"
Feb 26 12:19:36 akoww-a9mega ollama[349347]: [GIN] 2026/02/26 - 11:19:36 | 200 |   70.250091ms |       10.88.0.1 | POST     "/api/show"
Feb 26 12:19:36 akoww-a9mega ollama[349347]: [GIN] 2026/02/26 - 11:19:36 | 200 |   74.319276ms |       10.88.0.1 | POST     "/api/show"
Feb 26 12:19:36 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:36.883Z level=INFO source=server.go:431 msg="starting runner" cmd="/usr/bin/ollama runner --ollama-engine --port 41847"
Feb 26 12:19:36 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:36.931Z level=WARN source=cpu_linux.go:130 msg="failed to parse CPU allowed micro secs" error="strconv.ParseInt: parsing \"max\": invalid syntax"
Feb 26 12:19:36 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:36.976Z level=INFO source=server.go:247 msg="enabling flash attention"
Feb 26 12:19:36 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:36.976Z level=INFO source=server.go:431 msg="starting runner" cmd="/usr/bin/ollama runner --ollama-engine --model /root/.ollama/models/blobs/sha256-30e51a7cb1cf1333b9e298b90b4c7790fe2572d8736b002482a0ac96328a2ffb --port 42155"
Feb 26 12:19:36 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:36.976Z level=INFO source=sched.go:491 msg="system memory" total="124.0 GiB" free="119.4 GiB" free_swap="0 B"
Feb 26 12:19:36 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:36.976Z level=INFO source=sched.go:498 msg="gpu memory" id=00000000-c700-0000-0000-000000000000 library=Vulkan available="123.9 GiB" free="124.3 GiB" minimum="457.0 MiB" overhead="0 B"
Feb 26 12:19:36 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:36.976Z level=INFO source=server.go:757 msg="loading model" "model layers"=49 requested=-1
Feb 26 12:19:36 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:36.984Z level=INFO source=runner.go:1411 msg="starting ollama engine"
Feb 26 12:19:36 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:36.984Z level=INFO source=runner.go:1446 msg="Server listening on 127.0.0.1:42155"
Feb 26 12:19:36 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:36.988Z level=INFO source=runner.go:1284 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Enabled KvSize:262144 KvCacheType: NumThreads:16 GPULayers:49[ID:00000000-c700-0000-0000-000000000000 Layers:49(0..48)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
Feb 26 12:19:37 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:37.005Z level=INFO source=ggml.go:136 msg="" architecture=qwen3next file_type=Q4_K_M name="" description="" num_tensors=843 num_key_values=38
Feb 26 12:19:37 akoww-a9mega ollama[349347]: load_backend: loaded CPU backend from /usr/lib/ollama/libggml-cpu-icelake.so
Feb 26 12:19:37 akoww-a9mega ollama[349347]: ggml_vulkan: Found 1 Vulkan devices:
Feb 26 12:19:37 akoww-a9mega ollama[349347]: ggml_vulkan: 0 = Radeon 8060S Graphics (RADV GFX1151) (radv) | uma: 1 | fp16: 1 | bf16: 0 | warp size: 64 | shared memory: 65536 | int dot: 1 | matrix cores: KHR_coopmat
Feb 26 12:19:37 akoww-a9mega ollama[349347]: load_backend: loaded Vulkan backend from /usr/lib/ollama/vulkan/libggml-vulkan.so
Feb 26 12:19:37 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:37.034Z level=INFO source=ggml.go:104 msg=system CPU.0.SSE3=1 CPU.0.SSSE3=1 CPU.0.AVX=1 CPU.0.AVX2=1 CPU.0.F16C=1 CPU.0.FMA=1 CPU.0.BMI2=1 CPU.0.AVX512=1 CPU.0.AVX512_VBMI=1 CPU.0.AVX512_VNNI=1 CPU.0.LLAMAFILE=1 CPU.1.LLAMAFILE=1 compiler=cgo(gcc)
Feb 26 12:19:37 akoww-a9mega ollama[349347]: ggml_backend_vk_get_device_memory called: uuid 00000000-c700-0000-0000-000000000000
Feb 26 12:19:37 akoww-a9mega ollama[349347]: ggml_backend_vk_get_device_memory called: luid 0x0000000000000000
Feb 26 12:19:37 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:37.195Z level=INFO source=runner.go:1284 msg=load request="{Operation:alloc LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Enabled KvSize:262144 KvCacheType: NumThreads:16 GPULayers:49[ID:00000000-c700-0000-0000-000000000000 Layers:49(0..48)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
Feb 26 12:19:37 akoww-a9mega ollama[349347]: ggml_backend_vk_get_device_memory called: uuid 00000000-c700-0000-0000-000000000000
Feb 26 12:19:37 akoww-a9mega ollama[349347]: ggml_backend_vk_get_device_memory called: luid 0x0000000000000000
Feb 26 12:19:40 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:40.075Z level=INFO source=runner.go:1284 msg=load request="{Operation:commit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Enabled KvSize:262144 KvCacheType: NumThreads:16 GPULayers:49[ID:00000000-c700-0000-0000-000000000000 Layers:49(0..48)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
Feb 26 12:19:40 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:40.075Z level=INFO source=ggml.go:482 msg="offloading 48 repeating layers to GPU"
Feb 26 12:19:40 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:40.075Z level=INFO source=ggml.go:489 msg="offloading output layer to GPU"
Feb 26 12:19:40 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:40.075Z level=INFO source=ggml.go:494 msg="offloaded 49/49 layers to GPU"
Feb 26 12:19:40 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:40.076Z level=INFO source=device.go:240 msg="model weights" device=Vulkan0 size="48.0 GiB"
Feb 26 12:19:40 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:40.076Z level=INFO source=device.go:245 msg="model weights" device=CPU size="166.9 MiB"
Feb 26 12:19:40 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:40.076Z level=INFO source=device.go:251 msg="kv cache" device=Vulkan0 size="7.8 GiB"
Feb 26 12:19:40 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:40.076Z level=INFO source=device.go:262 msg="compute graph" device=Vulkan0 size="808.1 MiB"
Feb 26 12:19:40 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:40.076Z level=INFO source=device.go:267 msg="compute graph" device=CPU size="9.0 MiB"
Feb 26 12:19:40 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:40.076Z level=INFO source=device.go:272 msg="total memory" size="56.8 GiB"
Feb 26 12:19:40 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:40.076Z level=INFO source=sched.go:566 msg="loaded runners" count=1
Feb 26 12:19:40 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:40.076Z level=INFO source=server.go:1350 msg="waiting for llama runner to start responding"
Feb 26 12:19:40 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:40.076Z level=INFO source=server.go:1384 msg="waiting for server to become available" status="llm server loading model"
Feb 26 12:19:52 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:52.864Z level=INFO source=server.go:1388 msg="llama runner started in 15.89 seconds"
Feb 26 12:19:52 akoww-a9mega ollama[349347]: [GIN] 2026/02/26 - 11:19:52 | 200 | 16.060380064s |       10.88.0.1 | POST     "/api/generate"
Feb 26 12:19:58 akoww-a9mega ollama[349347]: SIGSEGV: segmentation violation
Feb 26 12:19:58 akoww-a9mega ollama[349347]: PC=0x7fe7c5e3957f m=17 sigcode=1 addr=0xb35900
Feb 26 12:19:58 akoww-a9mega ollama[349347]: signal arrived during cgo execution
Feb 26 12:19:58 akoww-a9mega ollama[349347]:
Feb 26 12:19:58 akoww-a9mega ollama[349347]: goroutine 1103 gp=0xc000102e00 m=17 mp=0xc00039a808 [syscall]:
Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime.cgocall(0x55a30dd6a330, 0xc0000baaa0)
Feb 26 12:19:58 akoww-a9mega ollama[349347]:         runtime/cgocall.go:167 +0x4b fp=0xc0000baa78 sp=0xc0000baa40 pc=0x55a30ce1ca6b
Feb 26 12:19:58 akoww-a9mega ollama[349347]: github.com/ollama/ollama/ml/backend/ggml._Cfunc_ggml_backend_sched_graph_compute_async(0x55a31cc045d0, 0x7fe754001be0)
Feb 26 12:19:58 akoww-a9mega ollama[349347]:         _cgo_gotypes.go:979 +0x4a fp=0xc0000baaa0 sp=0xc0000baa78 pc=0x55a30d307b0a
Feb 26 12:19:58 akoww-a9mega ollama[349347]: github.com/ollama/ollama/ml/backend/ggml.(*Context).ComputeWithNotify.func2(...)
Feb 26 12:19:58 akoww-a9mega ollama[349347]:         github.com/ollama/ollama/ml/backend/ggml/ggml.go:825
Feb 26 12:19:58 akoww-a9mega ollama[349347]: github.com/ollama/ollama/ml/backend/ggml.(*Context).ComputeWithNotify(0xc001afd380, 0xc0004bd5c0?, {0xc001255860, 0x1, 0x2?})
Feb 26 12:19:58 akoww-a9mega ollama[349347]:         github.com/ollama/ollama/ml/backend/ggml/ggml.go:825 +0x1b2 fp=0xc0000bab78 sp=0xc0000baaa0 pc=0x55a30d316492
Feb 26 12:19:58 akoww-a9mega ollama[349347]: github.com/ollama/ollama/runner/ollamarunner.(*Server).computeBatch(0xc0002af0e0, {0x0, {0x55a30e6c19d0, 0xc001afd380}, {0x55a30e6cee30, 0xc002443ae8}, {0xc00239d680, 0x9, 0x10}, {{0x55a30e6cee30, ...}, ...}, ...})
Feb 26 12:19:58 akoww-a9mega ollama[349347]:         github.com/ollama/ollama/runner/ollamarunner/runner.go:716 +0x862 fp=0xc0000baef0 sp=0xc0000bab78 pc=0x55a30d441282
Feb 26 12:19:58 akoww-a9mega ollama[349347]: github.com/ollama/ollama/runner/ollamarunner.(*Server).run.gowrap1()
Feb 26 12:19:58 akoww-a9mega ollama[349347]:         github.com/ollama/ollama/runner/ollamarunner/runner.go:459 +0x58 fp=0xc0000bafe0 sp=0xc0000baef0 pc=0x55a30d43ef78
Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime.goexit({})
Feb 26 12:19:58 akoww-a9mega ollama[349347]:         runtime/asm_amd64.s:1700 +0x1 fp=0xc0000bafe8 sp=0xc0000bafe0 pc=0x55a30ce27e61
Feb 26 12:19:58 akoww-a9mega ollama[349347]: created by github.com/ollama/ollama/runner/ollamarunner.(*Server).run in goroutine 54
Feb 26 12:19:58 akoww-a9mega ollama[349347]:         github.com/ollama/ollama/runner/ollamarunner/runner.go:459 +0x2cd
Feb 26 12:19:58 akoww-a9mega ollama[349347]:
Feb 26 12:19:58 akoww-a9mega ollama[349347]: goroutine 1 gp=0xc000002380 m=nil [IO wait]:
Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime.gopark(0x0?, 0x0?, 0x0?, 0x0?, 0x0?)
Feb 26 12:19:58 akoww-a9mega ollama[349347]:         runtime/proc.go:435 +0xce fp=0xc000b1b778 sp=0xc000b1b758 pc=0x55a30ce1feee
Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime.netpollblock(0xc0001177c8?, 0xcdb94a6?, 0xa3?)
Feb 26 12:19:58 akoww-a9mega ollama[349347]:         runtime/netpoll.go:575 +0xf7 fp=0xc000b1b7b0 sp=0xc000b1b778 pc=0x55a30cde5097
Feb 26 12:19:58 akoww-a9mega ollama[349347]: internal/poll.runtime_pollWait(0x7fe7d6426610, 0x72)
Feb 26 12:19:58 akoww-a9mega ollama[349347]:         runtime/netpoll.go:351 +0x85 fp=0xc000b1b7d0 sp=0xc000b1b7b0 pc=0x55a30ce1f105
Feb 26 12:19:58 akoww-a9mega ollama[349347]: internal/poll.(*pollDesc).wait(0xc00025f180?, 0x900000036?, 0x0)
Feb 26 12:19:58 akoww-a9mega ollama[349347]:         internal/poll/fd_poll_runtime.go:84 +0x27 fp=0xc000b1b7f8 sp=0xc000b1b7d0 pc=0x55a30cea7487
Feb 26 12:19:58 akoww-a9mega ollama[349347]: internal/poll.(*pollDesc).waitRead(...)
Feb 26 12:19:58 akoww-a9mega ollama[349347]:         internal/poll/fd_poll_runtime.go:89
Feb 26 12:19:58 akoww-a9mega ollama[349347]: internal/poll.(*FD).Accept(0xc00025f180)
Feb 26 12:19:58 akoww-a9mega ollama[349347]:         internal/poll/fd_unix.go:620 +0x295 fp=0xc000b1b8a0 sp=0xc000b1b7f8 pc=0x55a30ceac855
Feb 26 12:19:58 akoww-a9mega ollama[349347]: net.(*netFD).accept(0xc00025f180)
Feb 26 12:19:58 akoww-a9mega ollama[349347]:         net/fd_unix.go:172 +0x29 fp=0xc000b1b958 sp=0xc000b1b8a0 pc=0x55a30cf1fd49
Feb 26 12:19:58 akoww-a9mega ollama[349347]: net.(*TCPListener).accept(0xc00051fe40)
Feb 26 12:19:58 akoww-a9mega ollama[349347]:         net/tcpsock_posix.go:159 +0x1b fp=0xc000b1b9a8 sp=0xc000b1b958 pc=0x55a30cf35c5b
Feb 26 12:19:58 akoww-a9mega ollama[349347]: net.(*TCPListener).Accept(0xc00051fe40)
Feb 26 12:19:58 akoww-a9mega ollama[349347]:         net/tcpsock.go:380 +0x30 fp=0xc000b1b9d8 sp=0xc000b1b9a8 pc=0x55a30cf34b10
Feb 26 12:19:58 akoww-a9mega ollama[349347]: net/http.(*onceCloseListener).Accept(0xc0002be480?)
Feb 26 12:19:58 akoww-a9mega ollama[349347]:         <autogenerated>:1 +0x24 fp=0xc000b1b9f0 sp=0xc000b1b9d8 pc=0x55a30d14c9c4
Feb 26 12:19:58 akoww-a9mega ollama[349347]: net/http.(*Server).Serve(0xc000051600, {0x55a30e6b2bc0, 0xc00051fe40})
Feb 26 12:19:58 akoww-a9mega ollama[349347]:         net/http/server.go:3424 +0x30c fp=0xc000b1bb20 sp=0xc000b1b9f0 pc=0x55a30d12428c
Feb 26 12:19:58 akoww-a9mega ollama[349347]: github.com/ollama/ollama/runner/ollamarunner.Execute({0xc0001aa030, 0x4, 0x4})
Feb 26 12:19:58 akoww-a9mega ollama[349347]:         github.com/ollama/ollama/runner/ollamarunner/runner.go:1447 +0x94e fp=0xc000b1bcf0 sp=0xc000b1bb20 pc=0x55a30d44820e
Feb 26 12:19:58 akoww-a9mega ollama[349347]: github.com/ollama/ollama/runner.Execute({0xc0001aa010?, 0x0?, 0x0?})
Feb 26 12:19:58 akoww-a9mega ollama[349347]:         github.com/ollama/ollama/runner/runner.go:18 +0x10e fp=0xc000b1bd30 sp=0xc000b1bcf0 pc=0x55a30d4e776e
Feb 26 12:19:58 akoww-a9mega ollama[349347]: github.com/ollama/ollama/cmd.NewCLI.func3(0xc000051300?, {0x55a30e0d0236?, 0x4?, 0x55a30e0d023a?})
Feb 26 12:19:58 akoww-a9mega ollama[349347]:         github.com/ollama/ollama/cmd/cmd.go:2270 +0x45 fp=0xc000b1bd58 sp=0xc000b1bd30 pc=0x55a30dcfa845
Feb 26 12:19:58 akoww-a9mega ollama[349347]: github.com/spf13/cobra.(*Command).execute(0xc0002c1b08, {0xc000523ae0, 0x5, 0x5})
Feb 26 12:19:58 akoww-a9mega ollama[349347]:         github.com/spf13/cobra@v1.7.0/command.go:940 +0x85c fp=0xc000b1be78 sp=0xc000b1bd58 pc=0x55a30cf99cdc
Feb 26 12:19:58 akoww-a9mega ollama[349347]: github.com/spf13/cobra.(*Command).ExecuteC(0xc00001e908)
Feb 26 12:19:58 akoww-a9mega ollama[349347]:         github.com/spf13/cobra@v1.7.0/command.go:1068 +0x3a5 fp=0xc000b1bf30 sp=0xc000b1be78 pc=0x55a30cf9a525
Feb 26 12:19:58 akoww-a9mega ollama[349347]: github.com/spf13/cobra.(*Command).Execute(...)
Feb 26 12:19:58 akoww-a9mega ollama[349347]:         github.com/spf13/cobra@v1.7.0/command.go:992
Feb 26 12:19:58 akoww-a9mega ollama[349347]: github.com/spf13/cobra.(*Command).ExecuteContext(...)
Feb 26 12:19:58 akoww-a9mega ollama[349347]:         github.com/spf13/cobra@v1.7.0/command.go:985
Feb 26 12:19:58 akoww-a9mega ollama[349347]: main.main()
Feb 26 12:19:58 akoww-a9mega ollama[349347]:         github.com/ollama/ollama/main.go:12 +0x4d fp=0xc000b1bf50 sp=0xc000b1bf30 pc=0x55a30dcfcced
Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime.main()
Feb 26 12:19:58 akoww-a9mega ollama[349347]:         runtime/proc.go:283 +0x29d fp=0xc000b1bfe0 sp=0xc000b1bf50 pc=0x55a30cdec71d
Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime.goexit({})
Feb 26 12:19:58 akoww-a9mega ollama[349347]:         runtime/asm_amd64.s:1700 +0x1 fp=0xc000b1bfe8 sp=0xc000b1bfe0 pc=0x55a30ce27e61
Feb 26 12:19:58 akoww-a9mega ollama[349347]:
Feb 26 12:19:58 akoww-a9mega ollama[349347]: goroutine 2 gp=0xc000002e00 m=nil [force gc (idle)]:
Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime.gopark(0x0?, 0x0?, 0x0?, 0x0?, 0x0?)
Feb 26 12:19:58 akoww-a9mega ollama[349347]:         runtime/proc.go:435 +0xce fp=0xc0000aafa8 sp=0xc0000aaf88 pc=0x55a30ce1feee
Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime.goparkunlock(...)
Feb 26 12:19:58 akoww-a9mega ollama[349347]:         runtime/proc.go:441
Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime.forcegchelper()
Feb 26 12:19:58 akoww-a9mega ollama[349347]:         runtime/proc.go:348 +0xb8 fp=0xc0000aafe0 sp=0xc0000aafa8 pc=0x55a30cdeca58
Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime.goexit({})
Feb 26 12:19:58 akoww-a9mega ollama[349347]:         runtime/asm_amd64.s:1700 +0x1 fp=0xc0000aafe8 sp=0xc0000aafe0 pc=0x55a30ce27e61
Feb 26 12:19:58 akoww-a9mega ollama[349347]: created by runtime.init.7 in goroutine 1
Feb 26 12:19:58 akoww-a9mega ollama[349347]:         runtime/proc.go:336 +0x1a
Feb 26 12:19:58 akoww-a9mega ollama[349347]:
Feb 26 12:19:58 akoww-a9mega ollama[349347]: goroutine 3 gp=0xc000003340 m=nil [GC sweep wait]:
Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime.gopark(0x1?, 0x0?, 0x0?, 0x0?, 0x0?)
Feb 26 12:19:58 akoww-a9mega ollama[349347]:         runtime/proc.go:435 +0xce fp=0xc0000ab780 sp=0xc0000ab760 pc=0x55a30ce1feee
Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime.goparkunlock(...)
Feb 26 12:19:58 akoww-a9mega ollama[349347]:         runtime/proc.go:441
Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime.bgsweep(0xc0000d6000)
Feb 26 12:19:58 akoww-a9mega ollama[349347]:         runtime/mgcsweep.go:316 +0xdf fp=0xc0000ab7c8 sp=0xc0000ab780 pc=0x55a30cdd71ff
Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime.gcenable.gowrap1()
Feb 26 12:19:58 akoww-a9mega ollama[349347]:         runtime/mgc.go:204 +0x25 fp=0xc0000ab7e0 sp=0xc0000ab7c8 pc=0x55a30cdcb5e5
Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime.goexit({})
Feb 26 12:19:58 akoww-a9mega ollama[349347]:         runtime/asm_amd64.s:1700 +0x1 fp=0xc0000ab7e8 sp=0xc0000ab7e0 pc=0x55a30ce27e61
Feb 26 12:19:58 akoww-a9mega ollama[349347]: created by runtime.gcenable in goroutine 1
Feb 26 12:19:58 akoww-a9mega ollama[349347]:         runtime/mgc.go:204 +0x66
Feb 26 12:19:58 akoww-a9mega ollama[349347]:
Feb 26 12:19:58 akoww-a9mega ollama[349347]: goroutine 4 gp=0xc000003500 m=nil [GC scavenge wait]:
Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime.gopark(0x10000?, 0x55a30e2e6820?, 0x0?, 0x0?, 0x0?)
Feb 26 12:19:58 akoww-a9mega ollama[349347]:         runtime/proc.go:435 +0xce fp=0xc0000abf78 sp=0xc0000abf58 pc=0x55a30ce1feee
Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime.goparkunlock(...)
Feb 26 12:19:58 akoww-a9mega ollama[349347]:         runtime/proc.go:441
Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime.(*scavengerState).park(0x55a30f0e95a0)
Feb 26 12:19:58 akoww-a9mega ollama[349347]:         runtime/mgcscavenge.go:425 +0x49 fp=0xc0000abfa8 sp=0xc0000abf78 pc=0x55a30cdd4c49
Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime.bgscavenge(0xc0000d6000)
Feb 26 12:19:58 akoww-a9mega ollama[349347]:         runtime/mgcscavenge.go:658 +0x59 fp=0xc0000abfc8 sp=0xc0000abfa8 pc=0x55a30cdd51d9
Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime.gcenable.gowrap2()
Feb 26 12:19:58 akoww-a9mega ollama[349347]:         runtime/mgc.go:205 +0x25 fp=0xc0000abfe0 sp=0xc0000abfc8 pc=0x55a30cdcb585
Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime.goexit({})
Feb 26 12:19:58 akoww-a9mega ollama[349347]:         runtime/asm_amd64.s:1700 +0x1 fp=0xc0000abfe8 sp=0xc0000abfe0 pc=0x55a30ce27e61
Feb 26 12:19:58 akoww-a9mega ollama[349347]: created by runtime.gcenable in goroutine 1
Feb 26 12:19:58 akoww-a9mega ollama[349347]:         runtime/mgc.go:205 +0xa5
Feb 26 12:19:58 akoww-a9mega ollama[349347]:
Feb 26 12:19:58 akoww-a9mega ollama[349347]: goroutine 18 gp=0xc000186380 m=nil [finalizer wait]:
Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime.gopark(0x1b8?, 0xc000002380?, 0x1?, 0x23?, 0xc0000aa688?)
Feb 26 12:19:58 akoww-a9mega ollama[349347]:         runtime/proc.go:435 +0xce fp=0xc0000aa630 sp=0xc0000aa610 pc=0x55a30ce1feee
Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime.runfinq()
Feb 26 12:19:58 akoww-a9mega ollama[349347]:         runtime/mfinal.go:196 +0x107 fp=0xc0000aa7e0 sp=0xc0000aa630 pc=0x55a30cdca5a7
Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime.goexit({})
Feb 26 12:19:58 akoww-a9mega ollama[349347]:         runtime/asm_amd64.s:1700 +0x1 fp=0xc0000aa7e8 sp=0xc0000aa7e0 pc=0x55a30ce27e61
Feb 26 12:19:58 akoww-a9mega ollama[349347]: created by runtime.createfing in goroutine 1
Feb 26 12:19:58 akoww-a9mega ollama[349347]:         runtime/mfinal.go:166 +0x3d
Feb 26 12:19:58 akoww-a9mega ollama[349347]:
Feb 26 12:19:58 akoww-a9mega ollama[349347]: goroutine 19 gp=0xc000186e00 m=nil [chan receive]:
Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime.gopark(0xc0001ffcc0?, 0xc002b80018?, 0x60?, 0x67?, 0x55a30cf068a8?)
Feb 26 12:19:58 akoww-a9mega ollama[349347]:         runtime/proc.go:435 +0xce fp=0xc0000a6718 sp=0xc0000a66f8 pc=0x55a30ce1feee
Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime.chanrecv(0xc000182310, 0x0, 0x1)
Feb 26 12:19:58 akoww-a9mega ollama[349347]:         runtime/chan.go:664 +0x445 fp=0xc0000a6790 sp=0xc0000a6718 pc=0x55a30cdbc085
Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime.chanrecv1(0x0?, 0x0?)
Feb 26 12:19:58 akoww-a9mega ollama[349347]:         runtime/chan.go:506 +0x12 fp=0xc0000a67b8 sp=0xc0000a6790 pc=0x55a30cdbbc12
Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime.unique_runtime_registerUniqueMapCleanup.func2(...)
Feb 26 12:19:58 akoww-a9mega ollama[349347]:         runtime/mgc.go:1796
Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime.unique_runtime_registerUniqueMapCleanup.gowrap1()
Feb 26 12:19:58 akoww-a9mega ollama[349347]:         runtime/mgc.go:1799 +0x2f fp=0xc0000a67e0 sp=0xc0000a67b8 pc=0x55a30cdce78f
Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime.goexit({})
Feb 26 12:19:58 akoww-a9mega ollama[349347]:         runtime/asm_amd64.s:1700 +0x1 fp=0xc0000a67e8 sp=0xc0000a67e0 pc=0x55a30ce27e61
Feb 26 12:19:58 akoww-a9mega ollama[349347]: created by unique.runtime_registerUniqueMapCleanup in goroutine 1
Feb 26 12:19:58 akoww-a9mega ollama[349347]:         runtime/mgc.go:1794 +0x85
Feb 26 12:19:58 akoww-a9mega ollama[349347]:
 .....
Feb 26 12:19:58 akoww-a9mega ollama[349347]:
Feb 26 12:19:58 akoww-a9mega ollama[349347]: goroutine 54 gp=0xc000582fc0 m=nil [chan receive]:
Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime.gopark(0x30?, 0x55a30e5f0d80?, 0x1?, 0x0?, 0xc000b19798?)
Feb 26 12:19:58 akoww-a9mega ollama[349347]:         runtime/proc.go:435 +0xce fp=0xc000b19750 sp=0xc000b19730 pc=0x55a30ce1feee
Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime.chanrecv(0xc000dca070, 0x0, 0x1)
Feb 26 12:19:58 akoww-a9mega ollama[349347]:         runtime/chan.go:664 +0x445 fp=0xc000b197c8 sp=0xc000b19750 pc=0x55a30cdbc085
Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime.chanrecv1(0x55a30e113664?, 0x29?)
Feb 26 12:19:58 akoww-a9mega ollama[349347]:         runtime/chan.go:506 +0x12 fp=0xc000b197f0 sp=0xc000b197c8 pc=0x55a30cdbbc12
Feb 26 12:19:58 akoww-a9mega ollama[349347]: github.com/ollama/ollama/runner/ollamarunner.(*Server).forwardBatch(_, {0x1, {0x55a30e6c19d0, 0xc000dbb2c0}, {0x55a30e6cee30, 0xc0024823c0}, {0xc000db0358, 0x1, 0x1}, {{0x55a30e6cee30, ...}, ...}, ...})
Feb 26 12:19:58 akoww-a9mega ollama[349347]:         github.com/ollama/ollama/runner/ollamarunner/runner.go:476 +0xfa fp=0xc000b19b58 sp=0xc000b197f0 pc=0x55a30d43f09a
Feb 26 12:19:58 akoww-a9mega ollama[349347]: github.com/ollama/ollama/runner/ollamarunner.(*Server).run(0xc0002af0e0, {0x55a30e6b5520, 0xc000523b80})
Feb 26 12:19:58 akoww-a9mega ollama[349347]:         github.com/ollama/ollama/runner/ollamarunner/runner.go:453 +0x18c fp=0xc000b19fb8 sp=0xc000b19b58 pc=0x55a30d43ed4c
Feb 26 12:19:58 akoww-a9mega ollama[349347]: github.com/ollama/ollama/runner/ollamarunner.Execute.gowrap1()
Feb 26 12:19:58 akoww-a9mega ollama[349347]:         github.com/ollama/ollama/runner/ollamarunner/runner.go:1424 +0x28 fp=0xc000b19fe0 sp=0xc000b19fb8 pc=0x55a30d448488
Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime.goexit({})
Feb 26 12:19:58 akoww-a9mega ollama[349347]:         runtime/asm_amd64.s:1700 +0x1 fp=0xc000b19fe8 sp=0xc000b19fe0 pc=0x55a30ce27e61
Feb 26 12:19:58 akoww-a9mega ollama[349347]: created by github.com/ollama/ollama/runner/ollamarunner.Execute in goroutine 1
Feb 26 12:19:58 akoww-a9mega ollama[349347]:         github.com/ollama/ollama/runner/ollamarunner/runner.go:1424 +0x4c9
Feb 26 12:19:58 akoww-a9mega ollama[349347]:
Feb 26 12:19:58 akoww-a9mega ollama[349347]: goroutine 55 gp=0xc000583180 m=nil [select]:
Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime.gopark(0xc000049a08?, 0x2?, 0xc0?, 0x97?, 0xc00004986c?)
Feb 26 12:19:58 akoww-a9mega ollama[349347]:         runtime/proc.go:435 +0xce fp=0xc000049698 sp=0xc000049678 pc=0x55a30ce1feee
Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime.selectgo(0xc000049a08, 0xc000049868, 0x9?, 0x0, 0x1?, 0x1)
Feb 26 12:19:58 akoww-a9mega ollama[349347]:         runtime/select.go:351 +0x837 fp=0xc0000497d0 sp=0xc000049698 pc=0x55a30cdfec17
Feb 26 12:19:58 akoww-a9mega ollama[349347]: github.com/ollama/ollama/runner/ollamarunner.(*Server).completion(0xc0002af0e0, {0x55a30e6b2da0, 0xc0012301c0}, 0xc000424780)
Feb 26 12:19:58 akoww-a9mega ollama[349347]:         github.com/ollama/ollama/runner/ollamarunner/runner.go:956 +0xc4e fp=0xc000049ac0 sp=0xc0000497d0 pc=0x55a30d44352e
Feb 26 12:19:58 akoww-a9mega ollama[349347]: github.com/ollama/ollama/runner/ollamarunner.(*Server).completion-fm({0x55a30e6b2da0?, 0xc0012301c0?}, 0xc000049b40?)
Feb 26 12:19:58 akoww-a9mega ollama[349347]:         <autogenerated>:1 +0x36 fp=0xc000049af0 sp=0xc000049ac0 pc=0x55a30d448976
Feb 26 12:19:58 akoww-a9mega ollama[349347]: net/http.HandlerFunc.ServeHTTP(0xc000544780?, {0x55a30e6b2da0?, 0xc0012301c0?}, 0xc000049b60?)
Feb 26 12:19:58 akoww-a9mega ollama[349347]:         net/http/server.go:2294 +0x29 fp=0xc000049b18 sp=0xc000049af0 pc=0x55a30d1208c9
Feb 26 12:19:58 akoww-a9mega ollama[349347]: net/http.(*ServeMux).ServeHTTP(0x55a30cdc4ac5?, {0x55a30e6b2da0, 0xc0012301c0}, 0xc000424780)
Feb 26 12:19:58 akoww-a9mega ollama[349347]:         net/http/server.go:2822 +0x1c4 fp=0xc000049b68 sp=0xc000049b18 pc=0x55a30d1227c4
Feb 26 12:19:58 akoww-a9mega ollama[349347]: net/http.serverHandler.ServeHTTP({0x55a30e6af090?}, {0x55a30e6b2da0?, 0xc0012301c0?}, 0x1?)
Feb 26 12:19:58 akoww-a9mega ollama[349347]:         net/http/server.go:3301 +0x8e fp=0xc000049b98 sp=0xc000049b68 pc=0x55a30d14024e
Feb 26 12:19:58 akoww-a9mega ollama[349347]: net/http.(*conn).serve(0xc0002be480, {0x55a30e6b54e8, 0xc0002b96e0})
Feb 26 12:19:58 akoww-a9mega ollama[349347]:         net/http/server.go:2102 +0x625 fp=0xc000049fb8 sp=0xc000049b98 pc=0x55a30d11edc5
Feb 26 12:19:58 akoww-a9mega ollama[349347]: net/http.(*Server).Serve.gowrap3()
Feb 26 12:19:58 akoww-a9mega ollama[349347]:         net/http/server.go:3454 +0x28 fp=0xc000049fe0 sp=0xc000049fb8 pc=0x55a30d124688
Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime.goexit({})
Feb 26 12:19:58 akoww-a9mega ollama[349347]:         runtime/asm_amd64.s:1700 +0x1 fp=0xc000049fe8 sp=0xc000049fe0 pc=0x55a30ce27e61
Feb 26 12:19:58 akoww-a9mega ollama[349347]: created by net/http.(*Server).Serve in goroutine 1
Feb 26 12:19:58 akoww-a9mega ollama[349347]:         net/http/server.go:3454 +0x485
Feb 26 12:19:58 akoww-a9mega ollama[349347]:
Feb 26 12:19:58 akoww-a9mega ollama[349347]: goroutine 1102 gp=0xc0004cd180 m=nil [IO wait]:
Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime.gopark(0x463ddc00463dd800?, 0x463de400463de000?, 0x0?, 0xe8?, 0xb?)
Feb 26 12:19:58 akoww-a9mega ollama[349347]:         runtime/proc.go:435 +0xce fp=0xc00098bdd8 sp=0xc00098bdb8 pc=0x55a30ce1feee
Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime.netpollblock(0x55a30ce43798?, 0xcdb94a6?, 0xa3?)
Feb 26 12:19:58 akoww-a9mega ollama[349347]:         runtime/netpoll.go:575 +0xf7 fp=0xc00098be10 sp=0xc00098bdd8 pc=0x55a30cde5097
Feb 26 12:19:58 akoww-a9mega ollama[349347]: internal/poll.runtime_pollWait(0x7fe7d64264f8, 0x72)
Feb 26 12:19:58 akoww-a9mega ollama[349347]:         runtime/netpoll.go:351 +0x85 fp=0xc00098be30 sp=0xc00098be10 pc=0x55a30ce1f105
Feb 26 12:19:58 akoww-a9mega ollama[349347]: internal/poll.(*pollDesc).wait(0xc00025f200?, 0xc0002b97e1?, 0x0)
Feb 26 12:19:58 akoww-a9mega ollama[349347]:         internal/poll/fd_poll_runtime.go:84 +0x27 fp=0xc00098be58 sp=0xc00098be30 pc=0x55a30cea7487
Feb 26 12:19:58 akoww-a9mega ollama[349347]: internal/poll.(*pollDesc).waitRead(...)
Feb 26 12:19:58 akoww-a9mega ollama[349347]:         internal/poll/fd_poll_runtime.go:89
Feb 26 12:19:58 akoww-a9mega ollama[349347]: internal/poll.(*FD).Read(0xc00025f200, {0xc0002b97e1, 0x1, 0x1})
Feb 26 12:19:58 akoww-a9mega ollama[349347]:         internal/poll/fd_unix.go:165 +0x27a fp=0xc00098bef0 sp=0xc00098be58 pc=0x55a30cea877a
Feb 26 12:19:58 akoww-a9mega ollama[349347]: net.(*netFD).Read(0xc00025f200, {0xc0002b97e1?, 0xc00051ff18?, 0xc00098bf70?})
Feb 26 12:19:58 akoww-a9mega ollama[349347]:         net/fd_posix.go:55 +0x25 fp=0xc00098bf38 sp=0xc00098bef0 pc=0x55a30cf1dda5
Feb 26 12:19:58 akoww-a9mega ollama[349347]: net.(*conn).Read(0xc000154700, {0xc0002b97e1?, 0x463f9400463f9000?, 0x463f9c00463f9800?})
Feb 26 12:19:58 akoww-a9mega ollama[349347]:         net/net.go:194 +0x45 fp=0xc00098bf80 sp=0xc00098bf38 pc=0x55a30cf2c165
Feb 26 12:19:58 akoww-a9mega ollama[349347]: net/http.(*connReader).backgroundRead(0xc0002b97d0)
Feb 26 12:19:58 akoww-a9mega ollama[349347]:         net/http/server.go:690 +0x37 fp=0xc00098bfc8 sp=0xc00098bf80 pc=0x55a30d118c97
Feb 26 12:19:58 akoww-a9mega ollama[349347]: net/http.(*connReader).startBackgroundRead.gowrap2()
Feb 26 12:19:58 akoww-a9mega ollama[349347]:         net/http/server.go:686 +0x25 fp=0xc00098bfe0 sp=0xc00098bfc8 pc=0x55a30d118bc5
Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime.goexit({})
Feb 26 12:19:58 akoww-a9mega ollama[349347]:         runtime/asm_amd64.s:1700 +0x1 fp=0xc00098bfe8 sp=0xc00098bfe0 pc=0x55a30ce27e61
Feb 26 12:19:58 akoww-a9mega ollama[349347]: created by net/http.(*connReader).startBackgroundRead in goroutine 55
Feb 26 12:19:58 akoww-a9mega ollama[349347]:         net/http/server.go:686 +0xb6
Feb 26 12:19:58 akoww-a9mega ollama[349347]:
Feb 26 12:19:58 akoww-a9mega ollama[349347]: goroutine 971 gp=0xc0004cd880 m=nil [chan receive]:
Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime.gopark(0x30?, 0x55a30e5f0d80?, 0x1?, 0xc?, 0xc000ad2b20?)
Feb 26 12:19:58 akoww-a9mega ollama[349347]:         runtime/proc.go:435 +0xce fp=0xc000ad2ad8 sp=0xc000ad2ab8 pc=0x55a30ce1feee
Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime.chanrecv(0xc001aff5e0, 0x0, 0x1)
Feb 26 12:19:58 akoww-a9mega ollama[349347]:         runtime/chan.go:664 +0x445 fp=0xc000ad2b50 sp=0xc000ad2ad8 pc=0x55a30cdbc085
Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime.chanrecv1(0x55a30e117342?, 0x2c?)
Feb 26 12:19:58 akoww-a9mega ollama[349347]:         runtime/chan.go:506 +0x12 fp=0xc000ad2b78 sp=0xc000ad2b50 pc=0x55a30cdbbc12
Feb 26 12:19:58 akoww-a9mega ollama[349347]: github.com/ollama/ollama/runner/ollamarunner.(*Server).computeBatch(0xc0002af0e0, {0x1, {0x55a30e6c19d0, 0xc000dbb2c0}, {0x55a30e6cee30, 0xc0024823c0}, {0xc000db0358, 0x1, 0x1}, {{0x55a30e6cee30, ...}, ...}, ...})
Feb 26 12:19:58 akoww-a9mega ollama[349347]:         github.com/ollama/ollama/runner/ollamarunner/runner.go:645 +0x185 fp=0xc000ad2ef0 sp=0xc000ad2b78 pc=0x55a30d440ba5
Feb 26 12:19:58 akoww-a9mega ollama[349347]: github.com/ollama/ollama/runner/ollamarunner.(*Server).run.gowrap1()
Feb 26 12:19:58 akoww-a9mega ollama[349347]:         github.com/ollama/ollama/runner/ollamarunner/runner.go:459 +0x58 fp=0xc000ad2fe0 sp=0xc000ad2ef0 pc=0x55a30d43ef78
Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime.goexit({})
Feb 26 12:19:58 akoww-a9mega ollama[349347]:         runtime/asm_amd64.s:1700 +0x1 fp=0xc000ad2fe8 sp=0xc000ad2fe0 pc=0x55a30ce27e61
Feb 26 12:19:58 akoww-a9mega ollama[349347]: created by github.com/ollama/ollama/runner/ollamarunner.(*Server).run in goroutine 54
Feb 26 12:19:58 akoww-a9mega ollama[349347]:         github.com/ollama/ollama/runner/ollamarunner/runner.go:459 +0x2cd
Feb 26 12:19:58 akoww-a9mega ollama[349347]:
Feb 26 12:19:58 akoww-a9mega ollama[349347]: rax    0xb35900
Feb 26 12:19:58 akoww-a9mega ollama[349347]: rbx    0x0
Feb 26 12:19:58 akoww-a9mega ollama[349347]: rcx    0x0
Feb 26 12:19:58 akoww-a9mega ollama[349347]: rdx    0x7fe7bca41f00
Feb 26 12:19:58 akoww-a9mega ollama[349347]: rdi    0x200
Feb 26 12:19:58 akoww-a9mega ollama[349347]: rsi    0x80
Feb 26 12:19:58 akoww-a9mega ollama[349347]: rbp    0x7fe7c5b1c750
Feb 26 12:19:58 akoww-a9mega ollama[349347]: rsp    0x7fe7c5b1c680
Feb 26 12:19:58 akoww-a9mega ollama[349347]: r8     0x7fe7bca41f04
Feb 26 12:19:58 akoww-a9mega ollama[349347]: r9     0x80
Feb 26 12:19:58 akoww-a9mega ollama[349347]: r10    0x7f
Feb 26 12:19:58 akoww-a9mega ollama[349347]: r11    0x7f
Feb 26 12:19:58 akoww-a9mega ollama[349347]: r12    0x7fe7bca41f00
Feb 26 12:19:58 akoww-a9mega ollama[349347]: r13    0xb35900
Feb 26 12:19:58 akoww-a9mega ollama[349347]: r14    0x0
Feb 26 12:19:58 akoww-a9mega ollama[349347]: r15    0x40
Feb 26 12:19:58 akoww-a9mega ollama[349347]: rip    0x7fe7c5e3957f
Feb 26 12:19:58 akoww-a9mega ollama[349347]: rflags 0x10246
Feb 26 12:19:58 akoww-a9mega ollama[349347]: cs     0x33
Feb 26 12:19:58 akoww-a9mega ollama[349347]: fs     0x0
Feb 26 12:19:58 akoww-a9mega ollama[349347]: gs     0x0
Feb 26 12:19:58 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:58.176Z level=ERROR source=server.go:1610 msg="post predict" error="Post \"http://127.0.0.1:42155/completion\": EOF"
Feb 26 12:19:58 akoww-a9mega ollama[349347]: [GIN] 2026/02/26 - 11:19:58 | 500 |  123.372292ms |       10.88.0.1 | POST     "/api/chat"

OS

Linux

GPU

AMD

CPU

AMD

Ollama version

0.17.1

Originally created by @akoww on GitHub (Feb 26, 2026). Original GitHub issue: https://github.com/ollama/ollama/issues/14438 ### What is the issue? I downloaded the **qwen3-coder-next:latest** model and try to run it. loading into ram works but crashes on first request. I used the official **docker image 0.17.1** with Vulkan backend (CPU backend did also not work) on the other hand: qwen3-next:latest works perfectly fine. ### Relevant log output ```shell Feb 26 11:27:35 akoww-a9mega ollama[349347]: [GIN] 2026/02/26 - 10:27:35 | 200 | 30.938µs | 10.88.0.1 | HEAD "/" Feb 26 11:27:35 akoww-a9mega ollama[349347]: [GIN] 2026/02/26 - 10:27:35 | 200 | 62.652505ms | 10.88.0.1 | POST "/api/show" Feb 26 11:27:35 akoww-a9mega ollama[349347]: [GIN] 2026/02/26 - 10:27:35 | 200 | 65.47915ms | 10.88.0.1 | POST "/api/show" Feb 26 11:27:35 akoww-a9mega ollama[349347]: time=2026-02-26T10:27:35.971Z level=INFO source=server.go:431 msg="starting runner" cmd="/usr/bin/ollama runner --ollama-engine --port 45095" Feb 26 11:27:36 akoww-a9mega ollama[349347]: time=2026-02-26T10:27:36.014Z level=WARN source=cpu_linux.go:130 msg="failed to parse CPU allowed micro secs" error="strconv.ParseInt: parsing \"max\": invalid syntax" Feb 26 11:27:36 akoww-a9mega ollama[349347]: time=2026-02-26T10:27:36.061Z level=INFO source=server.go:247 msg="enabling flash attention" Feb 26 11:27:36 akoww-a9mega ollama[349347]: time=2026-02-26T10:27:36.061Z level=INFO source=server.go:431 msg="starting runner" cmd="/usr/bin/ollama runner --ollama-engine --model /root/.ollama/models/blobs/sha256-30e51a7cb1cf1333b9e298b90b4c7790fe2572d8736b002482a0ac96328a2ffb --port 34865" Feb 26 11:27:36 akoww-a9mega ollama[349347]: time=2026-02-26T10:27:36.062Z level=INFO source=sched.go:491 msg="system memory" total="124.0 GiB" free="119.4 GiB" free_swap="0 B" Feb 26 11:27:36 akoww-a9mega ollama[349347]: time=2026-02-26T10:27:36.062Z level=INFO source=sched.go:498 msg="gpu memory" id=00000000-c700-0000-0000-000000000000 library=Vulkan available="123.9 GiB" free="124.3 GiB" minimum="457.0 MiB" overhead="0 B" Feb 26 11:27:36 akoww-a9mega ollama[349347]: time=2026-02-26T10:27:36.062Z level=INFO source=server.go:757 msg="loading model" "model layers"=49 requested=-1 Feb 26 11:27:36 akoww-a9mega ollama[349347]: time=2026-02-26T10:27:36.068Z level=INFO source=runner.go:1411 msg="starting ollama engine" Feb 26 11:27:36 akoww-a9mega ollama[349347]: time=2026-02-26T10:27:36.068Z level=INFO source=runner.go:1446 msg="Server listening on 127.0.0.1:34865" Feb 26 11:27:36 akoww-a9mega ollama[349347]: time=2026-02-26T10:27:36.073Z level=INFO source=runner.go:1284 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Enabled KvSize:262144 KvCacheType: NumThreads:16 GPULayers:49[ID:00000000-c700-0000-0000-000000000000 Layers:49(0..48)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" Feb 26 11:27:36 akoww-a9mega ollama[349347]: time=2026-02-26T10:27:36.089Z level=INFO source=ggml.go:136 msg="" architecture=qwen3next file_type=Q4_K_M name="" description="" num_tensors=843 num_key_values=38 Feb 26 11:27:36 akoww-a9mega ollama[349347]: load_backend: loaded CPU backend from /usr/lib/ollama/libggml-cpu-icelake.so Feb 26 11:27:36 akoww-a9mega ollama[349347]: ggml_vulkan: Found 1 Vulkan devices: Feb 26 11:27:36 akoww-a9mega ollama[349347]: ggml_vulkan: 0 = Radeon 8060S Graphics (RADV GFX1151) (radv) | uma: 1 | fp16: 1 | bf16: 0 | warp size: 64 | shared memory: 65536 | int dot: 1 | matrix cores: KHR_coopmat Feb 26 11:27:36 akoww-a9mega ollama[349347]: load_backend: loaded Vulkan backend from /usr/lib/ollama/vulkan/libggml-vulkan.so Feb 26 11:27:36 akoww-a9mega ollama[349347]: time=2026-02-26T10:27:36.112Z level=INFO source=ggml.go:104 msg=system CPU.0.SSE3=1 CPU.0.SSSE3=1 CPU.0.AVX=1 CPU.0.AVX2=1 CPU.0.F16C=1 CPU.0.FMA=1 CPU.0.BMI2=1 CPU.0.AVX512=1 CPU.0.AVX512_VBMI=1 CPU.0.AVX512_VNNI=1 CPU.0.LLAMAFILE=1 CPU.1.LLAMAFILE=1 compiler=cgo(gcc) Feb 26 11:27:36 akoww-a9mega ollama[349347]: ggml_backend_vk_get_device_memory called: uuid 00000000-c700-0000-0000-000000000000 Feb 26 11:27:36 akoww-a9mega ollama[349347]: ggml_backend_vk_get_device_memory called: luid 0x0000000000000000 Feb 26 11:27:36 akoww-a9mega ollama[349347]: time=2026-02-26T10:27:36.272Z level=INFO source=runner.go:1284 msg=load request="{Operation:alloc LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Enabled KvSize:262144 KvCacheType: NumThreads:16 GPULayers:49[ID:00000000-c700-0000-0000-000000000000 Layers:49(0..48)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" Feb 26 11:27:36 akoww-a9mega ollama[349347]: ggml_backend_vk_get_device_memory called: uuid 00000000-c700-0000-0000-000000000000 Feb 26 11:27:36 akoww-a9mega ollama[349347]: ggml_backend_vk_get_device_memory called: luid 0x0000000000000000 Feb 26 11:27:39 akoww-a9mega ollama[349347]: time=2026-02-26T10:27:39.157Z level=INFO source=runner.go:1284 msg=load request="{Operation:commit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Enabled KvSize:262144 KvCacheType: NumThreads:16 GPULayers:49[ID:00000000-c700-0000-0000-000000000000 Layers:49(0..48)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" Feb 26 11:27:39 akoww-a9mega ollama[349347]: time=2026-02-26T10:27:39.157Z level=INFO source=ggml.go:482 msg="offloading 48 repeating layers to GPU" Feb 26 11:27:39 akoww-a9mega ollama[349347]: time=2026-02-26T10:27:39.157Z level=INFO source=ggml.go:489 msg="offloading output layer to GPU" Feb 26 11:27:39 akoww-a9mega ollama[349347]: time=2026-02-26T10:27:39.157Z level=INFO source=ggml.go:494 msg="offloaded 49/49 layers to GPU" Feb 26 11:27:39 akoww-a9mega ollama[349347]: time=2026-02-26T10:27:39.157Z level=INFO source=device.go:240 msg="model weights" device=Vulkan0 size="48.0 GiB" Feb 26 11:27:39 akoww-a9mega ollama[349347]: time=2026-02-26T10:27:39.157Z level=INFO source=device.go:245 msg="model weights" device=CPU size="166.9 MiB" Feb 26 11:27:39 akoww-a9mega ollama[349347]: time=2026-02-26T10:27:39.157Z level=INFO source=device.go:251 msg="kv cache" device=Vulkan0 size="7.8 GiB" Feb 26 11:27:39 akoww-a9mega ollama[349347]: time=2026-02-26T10:27:39.157Z level=INFO source=device.go:262 msg="compute graph" device=Vulkan0 size="808.1 MiB" Feb 26 11:27:39 akoww-a9mega ollama[349347]: time=2026-02-26T10:27:39.157Z level=INFO source=device.go:267 msg="compute graph" device=CPU size="9.0 MiB" Feb 26 11:27:39 akoww-a9mega ollama[349347]: time=2026-02-26T10:27:39.157Z level=INFO source=device.go:272 msg="total memory" size="56.8 GiB" Feb 26 11:27:39 akoww-a9mega ollama[349347]: time=2026-02-26T10:27:39.157Z level=INFO source=sched.go:566 msg="loaded runners" count=1 Feb 26 11:27:39 akoww-a9mega ollama[349347]: time=2026-02-26T10:27:39.157Z level=INFO source=server.go:1350 msg="waiting for llama runner to start responding" Feb 26 11:27:39 akoww-a9mega ollama[349347]: time=2026-02-26T10:27:39.157Z level=INFO source=server.go:1384 msg="waiting for server to become available" status="llm server loading model" Feb 26 11:27:51 akoww-a9mega ollama[349347]: time=2026-02-26T10:27:51.942Z level=INFO source=server.go:1388 msg="llama runner started in 15.88 seconds" Feb 26 11:27:51 akoww-a9mega ollama[349347]: [GIN] 2026/02/26 - 10:27:51 | 200 | 16.042613048s | 10.88.0.1 | POST "/api/generate" Feb 26 12:19:03 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:03.591Z level=INFO source=server.go:431 msg="starting runner" cmd="/usr/bin/ollama runner --ollama-engine --port 37115" Feb 26 12:19:03 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:03.639Z level=WARN source=cpu_linux.go:130 msg="failed to parse CPU allowed micro secs" error="strconv.ParseInt: parsing \"max\": invalid syntax" Feb 26 12:19:03 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:03.685Z level=INFO source=server.go:247 msg="enabling flash attention" Feb 26 12:19:03 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:03.685Z level=INFO source=server.go:431 msg="starting runner" cmd="/usr/bin/ollama runner --ollama-engine --model /root/.ollama/models/blobs/sha256-30e51a7cb1cf1333b9e298b90b4c7790fe2572d8736b002482a0ac96328a2ffb --port 38535" Feb 26 12:19:03 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:03.686Z level=INFO source=sched.go:491 msg="system memory" total="124.0 GiB" free="119.5 GiB" free_swap="0 B" Feb 26 12:19:03 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:03.686Z level=INFO source=sched.go:498 msg="gpu memory" id=00000000-c700-0000-0000-000000000000 library=Vulkan available="123.9 GiB" free="124.3 GiB" minimum="457.0 MiB" overhead="0 B" Feb 26 12:19:03 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:03.686Z level=INFO source=server.go:757 msg="loading model" "model layers"=49 requested=-1 Feb 26 12:19:03 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:03.694Z level=INFO source=runner.go:1411 msg="starting ollama engine" Feb 26 12:19:03 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:03.694Z level=INFO source=runner.go:1446 msg="Server listening on 127.0.0.1:38535" Feb 26 12:19:03 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:03.698Z level=INFO source=runner.go:1284 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Enabled KvSize:262144 KvCacheType: NumThreads:16 GPULayers:49[ID:00000000-c700-0000-0000-000000000000 Layers:49(0..48)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" Feb 26 12:19:03 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:03.716Z level=INFO source=ggml.go:136 msg="" architecture=qwen3next file_type=Q4_K_M name="" description="" num_tensors=843 num_key_values=38 Feb 26 12:19:03 akoww-a9mega ollama[349347]: load_backend: loaded CPU backend from /usr/lib/ollama/libggml-cpu-icelake.so Feb 26 12:19:03 akoww-a9mega ollama[349347]: ggml_vulkan: Found 1 Vulkan devices: Feb 26 12:19:03 akoww-a9mega ollama[349347]: ggml_vulkan: 0 = Radeon 8060S Graphics (RADV GFX1151) (radv) | uma: 1 | fp16: 1 | bf16: 0 | warp size: 64 | shared memory: 65536 | int dot: 1 | matrix cores: KHR_coopmat Feb 26 12:19:03 akoww-a9mega ollama[349347]: load_backend: loaded Vulkan backend from /usr/lib/ollama/vulkan/libggml-vulkan.so Feb 26 12:19:03 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:03.742Z level=INFO source=ggml.go:104 msg=system CPU.0.SSE3=1 CPU.0.SSSE3=1 CPU.0.AVX=1 CPU.0.AVX2=1 CPU.0.F16C=1 CPU.0.FMA=1 CPU.0.BMI2=1 CPU.0.AVX512=1 CPU.0.AVX512_VBMI=1 CPU.0.AVX512_VNNI=1 CPU.0.LLAMAFILE=1 CPU.1.LLAMAFILE=1 compiler=cgo(gcc) Feb 26 12:19:03 akoww-a9mega ollama[349347]: ggml_backend_vk_get_device_memory called: uuid 00000000-c700-0000-0000-000000000000 Feb 26 12:19:03 akoww-a9mega ollama[349347]: ggml_backend_vk_get_device_memory called: luid 0x0000000000000000 Feb 26 12:19:03 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:03.903Z level=INFO source=runner.go:1284 msg=load request="{Operation:alloc LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Enabled KvSize:262144 KvCacheType: NumThreads:16 GPULayers:49[ID:00000000-c700-0000-0000-000000000000 Layers:49(0..48)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" Feb 26 12:19:03 akoww-a9mega ollama[349347]: ggml_backend_vk_get_device_memory called: uuid 00000000-c700-0000-0000-000000000000 Feb 26 12:19:03 akoww-a9mega ollama[349347]: ggml_backend_vk_get_device_memory called: luid 0x0000000000000000 Feb 26 12:19:06 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:06.762Z level=INFO source=runner.go:1284 msg=load request="{Operation:commit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Enabled KvSize:262144 KvCacheType: NumThreads:16 GPULayers:49[ID:00000000-c700-0000-0000-000000000000 Layers:49(0..48)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" Feb 26 12:19:06 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:06.762Z level=INFO source=device.go:240 msg="model weights" device=Vulkan0 size="48.0 GiB" Feb 26 12:19:06 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:06.762Z level=INFO source=device.go:245 msg="model weights" device=CPU size="166.9 MiB" Feb 26 12:19:06 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:06.762Z level=INFO source=ggml.go:482 msg="offloading 48 repeating layers to GPU" Feb 26 12:19:06 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:06.762Z level=INFO source=ggml.go:489 msg="offloading output layer to GPU" Feb 26 12:19:06 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:06.762Z level=INFO source=ggml.go:494 msg="offloaded 49/49 layers to GPU" Feb 26 12:19:06 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:06.762Z level=INFO source=device.go:251 msg="kv cache" device=Vulkan0 size="7.8 GiB" Feb 26 12:19:06 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:06.762Z level=INFO source=device.go:262 msg="compute graph" device=Vulkan0 size="808.1 MiB" Feb 26 12:19:06 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:06.762Z level=INFO source=device.go:267 msg="compute graph" device=CPU size="9.0 MiB" Feb 26 12:19:06 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:06.762Z level=INFO source=device.go:272 msg="total memory" size="56.8 GiB" Feb 26 12:19:06 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:06.762Z level=INFO source=sched.go:566 msg="loaded runners" count=1 Feb 26 12:19:06 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:06.762Z level=INFO source=server.go:1350 msg="waiting for llama runner to start responding" Feb 26 12:19:06 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:06.763Z level=INFO source=server.go:1384 msg="waiting for server to become available" status="llm server loading model" Feb 26 12:19:09 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:09.269Z level=WARN source=server.go:1357 msg="client connection closed before server finished loading, aborting load" Feb 26 12:19:09 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:09.269Z level=ERROR source=sched.go:572 msg="error loading llama server" error="timed out waiting for llama runner to start: context canceled" Feb 26 12:19:09 akoww-a9mega ollama[349347]: [GIN] 2026/02/26 - 11:19:09 | 499 | 5.737795975s | 10.88.0.1 | POST "/api/chat" Feb 26 12:19:36 akoww-a9mega ollama[349347]: [GIN] 2026/02/26 - 11:19:36 | 200 | 49.803µs | 10.88.0.1 | HEAD "/" Feb 26 12:19:36 akoww-a9mega ollama[349347]: [GIN] 2026/02/26 - 11:19:36 | 200 | 70.250091ms | 10.88.0.1 | POST "/api/show" Feb 26 12:19:36 akoww-a9mega ollama[349347]: [GIN] 2026/02/26 - 11:19:36 | 200 | 74.319276ms | 10.88.0.1 | POST "/api/show" Feb 26 12:19:36 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:36.883Z level=INFO source=server.go:431 msg="starting runner" cmd="/usr/bin/ollama runner --ollama-engine --port 41847" Feb 26 12:19:36 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:36.931Z level=WARN source=cpu_linux.go:130 msg="failed to parse CPU allowed micro secs" error="strconv.ParseInt: parsing \"max\": invalid syntax" Feb 26 12:19:36 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:36.976Z level=INFO source=server.go:247 msg="enabling flash attention" Feb 26 12:19:36 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:36.976Z level=INFO source=server.go:431 msg="starting runner" cmd="/usr/bin/ollama runner --ollama-engine --model /root/.ollama/models/blobs/sha256-30e51a7cb1cf1333b9e298b90b4c7790fe2572d8736b002482a0ac96328a2ffb --port 42155" Feb 26 12:19:36 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:36.976Z level=INFO source=sched.go:491 msg="system memory" total="124.0 GiB" free="119.4 GiB" free_swap="0 B" Feb 26 12:19:36 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:36.976Z level=INFO source=sched.go:498 msg="gpu memory" id=00000000-c700-0000-0000-000000000000 library=Vulkan available="123.9 GiB" free="124.3 GiB" minimum="457.0 MiB" overhead="0 B" Feb 26 12:19:36 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:36.976Z level=INFO source=server.go:757 msg="loading model" "model layers"=49 requested=-1 Feb 26 12:19:36 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:36.984Z level=INFO source=runner.go:1411 msg="starting ollama engine" Feb 26 12:19:36 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:36.984Z level=INFO source=runner.go:1446 msg="Server listening on 127.0.0.1:42155" Feb 26 12:19:36 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:36.988Z level=INFO source=runner.go:1284 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Enabled KvSize:262144 KvCacheType: NumThreads:16 GPULayers:49[ID:00000000-c700-0000-0000-000000000000 Layers:49(0..48)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" Feb 26 12:19:37 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:37.005Z level=INFO source=ggml.go:136 msg="" architecture=qwen3next file_type=Q4_K_M name="" description="" num_tensors=843 num_key_values=38 Feb 26 12:19:37 akoww-a9mega ollama[349347]: load_backend: loaded CPU backend from /usr/lib/ollama/libggml-cpu-icelake.so Feb 26 12:19:37 akoww-a9mega ollama[349347]: ggml_vulkan: Found 1 Vulkan devices: Feb 26 12:19:37 akoww-a9mega ollama[349347]: ggml_vulkan: 0 = Radeon 8060S Graphics (RADV GFX1151) (radv) | uma: 1 | fp16: 1 | bf16: 0 | warp size: 64 | shared memory: 65536 | int dot: 1 | matrix cores: KHR_coopmat Feb 26 12:19:37 akoww-a9mega ollama[349347]: load_backend: loaded Vulkan backend from /usr/lib/ollama/vulkan/libggml-vulkan.so Feb 26 12:19:37 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:37.034Z level=INFO source=ggml.go:104 msg=system CPU.0.SSE3=1 CPU.0.SSSE3=1 CPU.0.AVX=1 CPU.0.AVX2=1 CPU.0.F16C=1 CPU.0.FMA=1 CPU.0.BMI2=1 CPU.0.AVX512=1 CPU.0.AVX512_VBMI=1 CPU.0.AVX512_VNNI=1 CPU.0.LLAMAFILE=1 CPU.1.LLAMAFILE=1 compiler=cgo(gcc) Feb 26 12:19:37 akoww-a9mega ollama[349347]: ggml_backend_vk_get_device_memory called: uuid 00000000-c700-0000-0000-000000000000 Feb 26 12:19:37 akoww-a9mega ollama[349347]: ggml_backend_vk_get_device_memory called: luid 0x0000000000000000 Feb 26 12:19:37 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:37.195Z level=INFO source=runner.go:1284 msg=load request="{Operation:alloc LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Enabled KvSize:262144 KvCacheType: NumThreads:16 GPULayers:49[ID:00000000-c700-0000-0000-000000000000 Layers:49(0..48)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" Feb 26 12:19:37 akoww-a9mega ollama[349347]: ggml_backend_vk_get_device_memory called: uuid 00000000-c700-0000-0000-000000000000 Feb 26 12:19:37 akoww-a9mega ollama[349347]: ggml_backend_vk_get_device_memory called: luid 0x0000000000000000 Feb 26 12:19:40 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:40.075Z level=INFO source=runner.go:1284 msg=load request="{Operation:commit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Enabled KvSize:262144 KvCacheType: NumThreads:16 GPULayers:49[ID:00000000-c700-0000-0000-000000000000 Layers:49(0..48)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" Feb 26 12:19:40 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:40.075Z level=INFO source=ggml.go:482 msg="offloading 48 repeating layers to GPU" Feb 26 12:19:40 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:40.075Z level=INFO source=ggml.go:489 msg="offloading output layer to GPU" Feb 26 12:19:40 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:40.075Z level=INFO source=ggml.go:494 msg="offloaded 49/49 layers to GPU" Feb 26 12:19:40 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:40.076Z level=INFO source=device.go:240 msg="model weights" device=Vulkan0 size="48.0 GiB" Feb 26 12:19:40 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:40.076Z level=INFO source=device.go:245 msg="model weights" device=CPU size="166.9 MiB" Feb 26 12:19:40 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:40.076Z level=INFO source=device.go:251 msg="kv cache" device=Vulkan0 size="7.8 GiB" Feb 26 12:19:40 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:40.076Z level=INFO source=device.go:262 msg="compute graph" device=Vulkan0 size="808.1 MiB" Feb 26 12:19:40 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:40.076Z level=INFO source=device.go:267 msg="compute graph" device=CPU size="9.0 MiB" Feb 26 12:19:40 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:40.076Z level=INFO source=device.go:272 msg="total memory" size="56.8 GiB" Feb 26 12:19:40 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:40.076Z level=INFO source=sched.go:566 msg="loaded runners" count=1 Feb 26 12:19:40 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:40.076Z level=INFO source=server.go:1350 msg="waiting for llama runner to start responding" Feb 26 12:19:40 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:40.076Z level=INFO source=server.go:1384 msg="waiting for server to become available" status="llm server loading model" Feb 26 12:19:52 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:52.864Z level=INFO source=server.go:1388 msg="llama runner started in 15.89 seconds" Feb 26 12:19:52 akoww-a9mega ollama[349347]: [GIN] 2026/02/26 - 11:19:52 | 200 | 16.060380064s | 10.88.0.1 | POST "/api/generate" Feb 26 12:19:58 akoww-a9mega ollama[349347]: SIGSEGV: segmentation violation Feb 26 12:19:58 akoww-a9mega ollama[349347]: PC=0x7fe7c5e3957f m=17 sigcode=1 addr=0xb35900 Feb 26 12:19:58 akoww-a9mega ollama[349347]: signal arrived during cgo execution Feb 26 12:19:58 akoww-a9mega ollama[349347]: Feb 26 12:19:58 akoww-a9mega ollama[349347]: goroutine 1103 gp=0xc000102e00 m=17 mp=0xc00039a808 [syscall]: Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime.cgocall(0x55a30dd6a330, 0xc0000baaa0) Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime/cgocall.go:167 +0x4b fp=0xc0000baa78 sp=0xc0000baa40 pc=0x55a30ce1ca6b Feb 26 12:19:58 akoww-a9mega ollama[349347]: github.com/ollama/ollama/ml/backend/ggml._Cfunc_ggml_backend_sched_graph_compute_async(0x55a31cc045d0, 0x7fe754001be0) Feb 26 12:19:58 akoww-a9mega ollama[349347]: _cgo_gotypes.go:979 +0x4a fp=0xc0000baaa0 sp=0xc0000baa78 pc=0x55a30d307b0a Feb 26 12:19:58 akoww-a9mega ollama[349347]: github.com/ollama/ollama/ml/backend/ggml.(*Context).ComputeWithNotify.func2(...) Feb 26 12:19:58 akoww-a9mega ollama[349347]: github.com/ollama/ollama/ml/backend/ggml/ggml.go:825 Feb 26 12:19:58 akoww-a9mega ollama[349347]: github.com/ollama/ollama/ml/backend/ggml.(*Context).ComputeWithNotify(0xc001afd380, 0xc0004bd5c0?, {0xc001255860, 0x1, 0x2?}) Feb 26 12:19:58 akoww-a9mega ollama[349347]: github.com/ollama/ollama/ml/backend/ggml/ggml.go:825 +0x1b2 fp=0xc0000bab78 sp=0xc0000baaa0 pc=0x55a30d316492 Feb 26 12:19:58 akoww-a9mega ollama[349347]: github.com/ollama/ollama/runner/ollamarunner.(*Server).computeBatch(0xc0002af0e0, {0x0, {0x55a30e6c19d0, 0xc001afd380}, {0x55a30e6cee30, 0xc002443ae8}, {0xc00239d680, 0x9, 0x10}, {{0x55a30e6cee30, ...}, ...}, ...}) Feb 26 12:19:58 akoww-a9mega ollama[349347]: github.com/ollama/ollama/runner/ollamarunner/runner.go:716 +0x862 fp=0xc0000baef0 sp=0xc0000bab78 pc=0x55a30d441282 Feb 26 12:19:58 akoww-a9mega ollama[349347]: github.com/ollama/ollama/runner/ollamarunner.(*Server).run.gowrap1() Feb 26 12:19:58 akoww-a9mega ollama[349347]: github.com/ollama/ollama/runner/ollamarunner/runner.go:459 +0x58 fp=0xc0000bafe0 sp=0xc0000baef0 pc=0x55a30d43ef78 Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime.goexit({}) Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime/asm_amd64.s:1700 +0x1 fp=0xc0000bafe8 sp=0xc0000bafe0 pc=0x55a30ce27e61 Feb 26 12:19:58 akoww-a9mega ollama[349347]: created by github.com/ollama/ollama/runner/ollamarunner.(*Server).run in goroutine 54 Feb 26 12:19:58 akoww-a9mega ollama[349347]: github.com/ollama/ollama/runner/ollamarunner/runner.go:459 +0x2cd Feb 26 12:19:58 akoww-a9mega ollama[349347]: Feb 26 12:19:58 akoww-a9mega ollama[349347]: goroutine 1 gp=0xc000002380 m=nil [IO wait]: Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime.gopark(0x0?, 0x0?, 0x0?, 0x0?, 0x0?) Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime/proc.go:435 +0xce fp=0xc000b1b778 sp=0xc000b1b758 pc=0x55a30ce1feee Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime.netpollblock(0xc0001177c8?, 0xcdb94a6?, 0xa3?) Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime/netpoll.go:575 +0xf7 fp=0xc000b1b7b0 sp=0xc000b1b778 pc=0x55a30cde5097 Feb 26 12:19:58 akoww-a9mega ollama[349347]: internal/poll.runtime_pollWait(0x7fe7d6426610, 0x72) Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime/netpoll.go:351 +0x85 fp=0xc000b1b7d0 sp=0xc000b1b7b0 pc=0x55a30ce1f105 Feb 26 12:19:58 akoww-a9mega ollama[349347]: internal/poll.(*pollDesc).wait(0xc00025f180?, 0x900000036?, 0x0) Feb 26 12:19:58 akoww-a9mega ollama[349347]: internal/poll/fd_poll_runtime.go:84 +0x27 fp=0xc000b1b7f8 sp=0xc000b1b7d0 pc=0x55a30cea7487 Feb 26 12:19:58 akoww-a9mega ollama[349347]: internal/poll.(*pollDesc).waitRead(...) Feb 26 12:19:58 akoww-a9mega ollama[349347]: internal/poll/fd_poll_runtime.go:89 Feb 26 12:19:58 akoww-a9mega ollama[349347]: internal/poll.(*FD).Accept(0xc00025f180) Feb 26 12:19:58 akoww-a9mega ollama[349347]: internal/poll/fd_unix.go:620 +0x295 fp=0xc000b1b8a0 sp=0xc000b1b7f8 pc=0x55a30ceac855 Feb 26 12:19:58 akoww-a9mega ollama[349347]: net.(*netFD).accept(0xc00025f180) Feb 26 12:19:58 akoww-a9mega ollama[349347]: net/fd_unix.go:172 +0x29 fp=0xc000b1b958 sp=0xc000b1b8a0 pc=0x55a30cf1fd49 Feb 26 12:19:58 akoww-a9mega ollama[349347]: net.(*TCPListener).accept(0xc00051fe40) Feb 26 12:19:58 akoww-a9mega ollama[349347]: net/tcpsock_posix.go:159 +0x1b fp=0xc000b1b9a8 sp=0xc000b1b958 pc=0x55a30cf35c5b Feb 26 12:19:58 akoww-a9mega ollama[349347]: net.(*TCPListener).Accept(0xc00051fe40) Feb 26 12:19:58 akoww-a9mega ollama[349347]: net/tcpsock.go:380 +0x30 fp=0xc000b1b9d8 sp=0xc000b1b9a8 pc=0x55a30cf34b10 Feb 26 12:19:58 akoww-a9mega ollama[349347]: net/http.(*onceCloseListener).Accept(0xc0002be480?) Feb 26 12:19:58 akoww-a9mega ollama[349347]: <autogenerated>:1 +0x24 fp=0xc000b1b9f0 sp=0xc000b1b9d8 pc=0x55a30d14c9c4 Feb 26 12:19:58 akoww-a9mega ollama[349347]: net/http.(*Server).Serve(0xc000051600, {0x55a30e6b2bc0, 0xc00051fe40}) Feb 26 12:19:58 akoww-a9mega ollama[349347]: net/http/server.go:3424 +0x30c fp=0xc000b1bb20 sp=0xc000b1b9f0 pc=0x55a30d12428c Feb 26 12:19:58 akoww-a9mega ollama[349347]: github.com/ollama/ollama/runner/ollamarunner.Execute({0xc0001aa030, 0x4, 0x4}) Feb 26 12:19:58 akoww-a9mega ollama[349347]: github.com/ollama/ollama/runner/ollamarunner/runner.go:1447 +0x94e fp=0xc000b1bcf0 sp=0xc000b1bb20 pc=0x55a30d44820e Feb 26 12:19:58 akoww-a9mega ollama[349347]: github.com/ollama/ollama/runner.Execute({0xc0001aa010?, 0x0?, 0x0?}) Feb 26 12:19:58 akoww-a9mega ollama[349347]: github.com/ollama/ollama/runner/runner.go:18 +0x10e fp=0xc000b1bd30 sp=0xc000b1bcf0 pc=0x55a30d4e776e Feb 26 12:19:58 akoww-a9mega ollama[349347]: github.com/ollama/ollama/cmd.NewCLI.func3(0xc000051300?, {0x55a30e0d0236?, 0x4?, 0x55a30e0d023a?}) Feb 26 12:19:58 akoww-a9mega ollama[349347]: github.com/ollama/ollama/cmd/cmd.go:2270 +0x45 fp=0xc000b1bd58 sp=0xc000b1bd30 pc=0x55a30dcfa845 Feb 26 12:19:58 akoww-a9mega ollama[349347]: github.com/spf13/cobra.(*Command).execute(0xc0002c1b08, {0xc000523ae0, 0x5, 0x5}) Feb 26 12:19:58 akoww-a9mega ollama[349347]: github.com/spf13/cobra@v1.7.0/command.go:940 +0x85c fp=0xc000b1be78 sp=0xc000b1bd58 pc=0x55a30cf99cdc Feb 26 12:19:58 akoww-a9mega ollama[349347]: github.com/spf13/cobra.(*Command).ExecuteC(0xc00001e908) Feb 26 12:19:58 akoww-a9mega ollama[349347]: github.com/spf13/cobra@v1.7.0/command.go:1068 +0x3a5 fp=0xc000b1bf30 sp=0xc000b1be78 pc=0x55a30cf9a525 Feb 26 12:19:58 akoww-a9mega ollama[349347]: github.com/spf13/cobra.(*Command).Execute(...) Feb 26 12:19:58 akoww-a9mega ollama[349347]: github.com/spf13/cobra@v1.7.0/command.go:992 Feb 26 12:19:58 akoww-a9mega ollama[349347]: github.com/spf13/cobra.(*Command).ExecuteContext(...) Feb 26 12:19:58 akoww-a9mega ollama[349347]: github.com/spf13/cobra@v1.7.0/command.go:985 Feb 26 12:19:58 akoww-a9mega ollama[349347]: main.main() Feb 26 12:19:58 akoww-a9mega ollama[349347]: github.com/ollama/ollama/main.go:12 +0x4d fp=0xc000b1bf50 sp=0xc000b1bf30 pc=0x55a30dcfcced Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime.main() Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime/proc.go:283 +0x29d fp=0xc000b1bfe0 sp=0xc000b1bf50 pc=0x55a30cdec71d Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime.goexit({}) Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime/asm_amd64.s:1700 +0x1 fp=0xc000b1bfe8 sp=0xc000b1bfe0 pc=0x55a30ce27e61 Feb 26 12:19:58 akoww-a9mega ollama[349347]: Feb 26 12:19:58 akoww-a9mega ollama[349347]: goroutine 2 gp=0xc000002e00 m=nil [force gc (idle)]: Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime.gopark(0x0?, 0x0?, 0x0?, 0x0?, 0x0?) Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime/proc.go:435 +0xce fp=0xc0000aafa8 sp=0xc0000aaf88 pc=0x55a30ce1feee Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime.goparkunlock(...) Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime/proc.go:441 Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime.forcegchelper() Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime/proc.go:348 +0xb8 fp=0xc0000aafe0 sp=0xc0000aafa8 pc=0x55a30cdeca58 Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime.goexit({}) Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime/asm_amd64.s:1700 +0x1 fp=0xc0000aafe8 sp=0xc0000aafe0 pc=0x55a30ce27e61 Feb 26 12:19:58 akoww-a9mega ollama[349347]: created by runtime.init.7 in goroutine 1 Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime/proc.go:336 +0x1a Feb 26 12:19:58 akoww-a9mega ollama[349347]: Feb 26 12:19:58 akoww-a9mega ollama[349347]: goroutine 3 gp=0xc000003340 m=nil [GC sweep wait]: Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime.gopark(0x1?, 0x0?, 0x0?, 0x0?, 0x0?) Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime/proc.go:435 +0xce fp=0xc0000ab780 sp=0xc0000ab760 pc=0x55a30ce1feee Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime.goparkunlock(...) Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime/proc.go:441 Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime.bgsweep(0xc0000d6000) Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime/mgcsweep.go:316 +0xdf fp=0xc0000ab7c8 sp=0xc0000ab780 pc=0x55a30cdd71ff Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime.gcenable.gowrap1() Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime/mgc.go:204 +0x25 fp=0xc0000ab7e0 sp=0xc0000ab7c8 pc=0x55a30cdcb5e5 Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime.goexit({}) Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime/asm_amd64.s:1700 +0x1 fp=0xc0000ab7e8 sp=0xc0000ab7e0 pc=0x55a30ce27e61 Feb 26 12:19:58 akoww-a9mega ollama[349347]: created by runtime.gcenable in goroutine 1 Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime/mgc.go:204 +0x66 Feb 26 12:19:58 akoww-a9mega ollama[349347]: Feb 26 12:19:58 akoww-a9mega ollama[349347]: goroutine 4 gp=0xc000003500 m=nil [GC scavenge wait]: Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime.gopark(0x10000?, 0x55a30e2e6820?, 0x0?, 0x0?, 0x0?) Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime/proc.go:435 +0xce fp=0xc0000abf78 sp=0xc0000abf58 pc=0x55a30ce1feee Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime.goparkunlock(...) Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime/proc.go:441 Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime.(*scavengerState).park(0x55a30f0e95a0) Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime/mgcscavenge.go:425 +0x49 fp=0xc0000abfa8 sp=0xc0000abf78 pc=0x55a30cdd4c49 Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime.bgscavenge(0xc0000d6000) Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime/mgcscavenge.go:658 +0x59 fp=0xc0000abfc8 sp=0xc0000abfa8 pc=0x55a30cdd51d9 Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime.gcenable.gowrap2() Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime/mgc.go:205 +0x25 fp=0xc0000abfe0 sp=0xc0000abfc8 pc=0x55a30cdcb585 Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime.goexit({}) Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime/asm_amd64.s:1700 +0x1 fp=0xc0000abfe8 sp=0xc0000abfe0 pc=0x55a30ce27e61 Feb 26 12:19:58 akoww-a9mega ollama[349347]: created by runtime.gcenable in goroutine 1 Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime/mgc.go:205 +0xa5 Feb 26 12:19:58 akoww-a9mega ollama[349347]: Feb 26 12:19:58 akoww-a9mega ollama[349347]: goroutine 18 gp=0xc000186380 m=nil [finalizer wait]: Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime.gopark(0x1b8?, 0xc000002380?, 0x1?, 0x23?, 0xc0000aa688?) Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime/proc.go:435 +0xce fp=0xc0000aa630 sp=0xc0000aa610 pc=0x55a30ce1feee Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime.runfinq() Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime/mfinal.go:196 +0x107 fp=0xc0000aa7e0 sp=0xc0000aa630 pc=0x55a30cdca5a7 Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime.goexit({}) Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime/asm_amd64.s:1700 +0x1 fp=0xc0000aa7e8 sp=0xc0000aa7e0 pc=0x55a30ce27e61 Feb 26 12:19:58 akoww-a9mega ollama[349347]: created by runtime.createfing in goroutine 1 Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime/mfinal.go:166 +0x3d Feb 26 12:19:58 akoww-a9mega ollama[349347]: Feb 26 12:19:58 akoww-a9mega ollama[349347]: goroutine 19 gp=0xc000186e00 m=nil [chan receive]: Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime.gopark(0xc0001ffcc0?, 0xc002b80018?, 0x60?, 0x67?, 0x55a30cf068a8?) Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime/proc.go:435 +0xce fp=0xc0000a6718 sp=0xc0000a66f8 pc=0x55a30ce1feee Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime.chanrecv(0xc000182310, 0x0, 0x1) Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime/chan.go:664 +0x445 fp=0xc0000a6790 sp=0xc0000a6718 pc=0x55a30cdbc085 Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime.chanrecv1(0x0?, 0x0?) Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime/chan.go:506 +0x12 fp=0xc0000a67b8 sp=0xc0000a6790 pc=0x55a30cdbbc12 Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime.unique_runtime_registerUniqueMapCleanup.func2(...) Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime/mgc.go:1796 Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime.unique_runtime_registerUniqueMapCleanup.gowrap1() Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime/mgc.go:1799 +0x2f fp=0xc0000a67e0 sp=0xc0000a67b8 pc=0x55a30cdce78f Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime.goexit({}) Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime/asm_amd64.s:1700 +0x1 fp=0xc0000a67e8 sp=0xc0000a67e0 pc=0x55a30ce27e61 Feb 26 12:19:58 akoww-a9mega ollama[349347]: created by unique.runtime_registerUniqueMapCleanup in goroutine 1 Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime/mgc.go:1794 +0x85 Feb 26 12:19:58 akoww-a9mega ollama[349347]: ..... Feb 26 12:19:58 akoww-a9mega ollama[349347]: Feb 26 12:19:58 akoww-a9mega ollama[349347]: goroutine 54 gp=0xc000582fc0 m=nil [chan receive]: Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime.gopark(0x30?, 0x55a30e5f0d80?, 0x1?, 0x0?, 0xc000b19798?) Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime/proc.go:435 +0xce fp=0xc000b19750 sp=0xc000b19730 pc=0x55a30ce1feee Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime.chanrecv(0xc000dca070, 0x0, 0x1) Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime/chan.go:664 +0x445 fp=0xc000b197c8 sp=0xc000b19750 pc=0x55a30cdbc085 Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime.chanrecv1(0x55a30e113664?, 0x29?) Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime/chan.go:506 +0x12 fp=0xc000b197f0 sp=0xc000b197c8 pc=0x55a30cdbbc12 Feb 26 12:19:58 akoww-a9mega ollama[349347]: github.com/ollama/ollama/runner/ollamarunner.(*Server).forwardBatch(_, {0x1, {0x55a30e6c19d0, 0xc000dbb2c0}, {0x55a30e6cee30, 0xc0024823c0}, {0xc000db0358, 0x1, 0x1}, {{0x55a30e6cee30, ...}, ...}, ...}) Feb 26 12:19:58 akoww-a9mega ollama[349347]: github.com/ollama/ollama/runner/ollamarunner/runner.go:476 +0xfa fp=0xc000b19b58 sp=0xc000b197f0 pc=0x55a30d43f09a Feb 26 12:19:58 akoww-a9mega ollama[349347]: github.com/ollama/ollama/runner/ollamarunner.(*Server).run(0xc0002af0e0, {0x55a30e6b5520, 0xc000523b80}) Feb 26 12:19:58 akoww-a9mega ollama[349347]: github.com/ollama/ollama/runner/ollamarunner/runner.go:453 +0x18c fp=0xc000b19fb8 sp=0xc000b19b58 pc=0x55a30d43ed4c Feb 26 12:19:58 akoww-a9mega ollama[349347]: github.com/ollama/ollama/runner/ollamarunner.Execute.gowrap1() Feb 26 12:19:58 akoww-a9mega ollama[349347]: github.com/ollama/ollama/runner/ollamarunner/runner.go:1424 +0x28 fp=0xc000b19fe0 sp=0xc000b19fb8 pc=0x55a30d448488 Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime.goexit({}) Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime/asm_amd64.s:1700 +0x1 fp=0xc000b19fe8 sp=0xc000b19fe0 pc=0x55a30ce27e61 Feb 26 12:19:58 akoww-a9mega ollama[349347]: created by github.com/ollama/ollama/runner/ollamarunner.Execute in goroutine 1 Feb 26 12:19:58 akoww-a9mega ollama[349347]: github.com/ollama/ollama/runner/ollamarunner/runner.go:1424 +0x4c9 Feb 26 12:19:58 akoww-a9mega ollama[349347]: Feb 26 12:19:58 akoww-a9mega ollama[349347]: goroutine 55 gp=0xc000583180 m=nil [select]: Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime.gopark(0xc000049a08?, 0x2?, 0xc0?, 0x97?, 0xc00004986c?) Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime/proc.go:435 +0xce fp=0xc000049698 sp=0xc000049678 pc=0x55a30ce1feee Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime.selectgo(0xc000049a08, 0xc000049868, 0x9?, 0x0, 0x1?, 0x1) Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime/select.go:351 +0x837 fp=0xc0000497d0 sp=0xc000049698 pc=0x55a30cdfec17 Feb 26 12:19:58 akoww-a9mega ollama[349347]: github.com/ollama/ollama/runner/ollamarunner.(*Server).completion(0xc0002af0e0, {0x55a30e6b2da0, 0xc0012301c0}, 0xc000424780) Feb 26 12:19:58 akoww-a9mega ollama[349347]: github.com/ollama/ollama/runner/ollamarunner/runner.go:956 +0xc4e fp=0xc000049ac0 sp=0xc0000497d0 pc=0x55a30d44352e Feb 26 12:19:58 akoww-a9mega ollama[349347]: github.com/ollama/ollama/runner/ollamarunner.(*Server).completion-fm({0x55a30e6b2da0?, 0xc0012301c0?}, 0xc000049b40?) Feb 26 12:19:58 akoww-a9mega ollama[349347]: <autogenerated>:1 +0x36 fp=0xc000049af0 sp=0xc000049ac0 pc=0x55a30d448976 Feb 26 12:19:58 akoww-a9mega ollama[349347]: net/http.HandlerFunc.ServeHTTP(0xc000544780?, {0x55a30e6b2da0?, 0xc0012301c0?}, 0xc000049b60?) Feb 26 12:19:58 akoww-a9mega ollama[349347]: net/http/server.go:2294 +0x29 fp=0xc000049b18 sp=0xc000049af0 pc=0x55a30d1208c9 Feb 26 12:19:58 akoww-a9mega ollama[349347]: net/http.(*ServeMux).ServeHTTP(0x55a30cdc4ac5?, {0x55a30e6b2da0, 0xc0012301c0}, 0xc000424780) Feb 26 12:19:58 akoww-a9mega ollama[349347]: net/http/server.go:2822 +0x1c4 fp=0xc000049b68 sp=0xc000049b18 pc=0x55a30d1227c4 Feb 26 12:19:58 akoww-a9mega ollama[349347]: net/http.serverHandler.ServeHTTP({0x55a30e6af090?}, {0x55a30e6b2da0?, 0xc0012301c0?}, 0x1?) Feb 26 12:19:58 akoww-a9mega ollama[349347]: net/http/server.go:3301 +0x8e fp=0xc000049b98 sp=0xc000049b68 pc=0x55a30d14024e Feb 26 12:19:58 akoww-a9mega ollama[349347]: net/http.(*conn).serve(0xc0002be480, {0x55a30e6b54e8, 0xc0002b96e0}) Feb 26 12:19:58 akoww-a9mega ollama[349347]: net/http/server.go:2102 +0x625 fp=0xc000049fb8 sp=0xc000049b98 pc=0x55a30d11edc5 Feb 26 12:19:58 akoww-a9mega ollama[349347]: net/http.(*Server).Serve.gowrap3() Feb 26 12:19:58 akoww-a9mega ollama[349347]: net/http/server.go:3454 +0x28 fp=0xc000049fe0 sp=0xc000049fb8 pc=0x55a30d124688 Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime.goexit({}) Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime/asm_amd64.s:1700 +0x1 fp=0xc000049fe8 sp=0xc000049fe0 pc=0x55a30ce27e61 Feb 26 12:19:58 akoww-a9mega ollama[349347]: created by net/http.(*Server).Serve in goroutine 1 Feb 26 12:19:58 akoww-a9mega ollama[349347]: net/http/server.go:3454 +0x485 Feb 26 12:19:58 akoww-a9mega ollama[349347]: Feb 26 12:19:58 akoww-a9mega ollama[349347]: goroutine 1102 gp=0xc0004cd180 m=nil [IO wait]: Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime.gopark(0x463ddc00463dd800?, 0x463de400463de000?, 0x0?, 0xe8?, 0xb?) Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime/proc.go:435 +0xce fp=0xc00098bdd8 sp=0xc00098bdb8 pc=0x55a30ce1feee Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime.netpollblock(0x55a30ce43798?, 0xcdb94a6?, 0xa3?) Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime/netpoll.go:575 +0xf7 fp=0xc00098be10 sp=0xc00098bdd8 pc=0x55a30cde5097 Feb 26 12:19:58 akoww-a9mega ollama[349347]: internal/poll.runtime_pollWait(0x7fe7d64264f8, 0x72) Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime/netpoll.go:351 +0x85 fp=0xc00098be30 sp=0xc00098be10 pc=0x55a30ce1f105 Feb 26 12:19:58 akoww-a9mega ollama[349347]: internal/poll.(*pollDesc).wait(0xc00025f200?, 0xc0002b97e1?, 0x0) Feb 26 12:19:58 akoww-a9mega ollama[349347]: internal/poll/fd_poll_runtime.go:84 +0x27 fp=0xc00098be58 sp=0xc00098be30 pc=0x55a30cea7487 Feb 26 12:19:58 akoww-a9mega ollama[349347]: internal/poll.(*pollDesc).waitRead(...) Feb 26 12:19:58 akoww-a9mega ollama[349347]: internal/poll/fd_poll_runtime.go:89 Feb 26 12:19:58 akoww-a9mega ollama[349347]: internal/poll.(*FD).Read(0xc00025f200, {0xc0002b97e1, 0x1, 0x1}) Feb 26 12:19:58 akoww-a9mega ollama[349347]: internal/poll/fd_unix.go:165 +0x27a fp=0xc00098bef0 sp=0xc00098be58 pc=0x55a30cea877a Feb 26 12:19:58 akoww-a9mega ollama[349347]: net.(*netFD).Read(0xc00025f200, {0xc0002b97e1?, 0xc00051ff18?, 0xc00098bf70?}) Feb 26 12:19:58 akoww-a9mega ollama[349347]: net/fd_posix.go:55 +0x25 fp=0xc00098bf38 sp=0xc00098bef0 pc=0x55a30cf1dda5 Feb 26 12:19:58 akoww-a9mega ollama[349347]: net.(*conn).Read(0xc000154700, {0xc0002b97e1?, 0x463f9400463f9000?, 0x463f9c00463f9800?}) Feb 26 12:19:58 akoww-a9mega ollama[349347]: net/net.go:194 +0x45 fp=0xc00098bf80 sp=0xc00098bf38 pc=0x55a30cf2c165 Feb 26 12:19:58 akoww-a9mega ollama[349347]: net/http.(*connReader).backgroundRead(0xc0002b97d0) Feb 26 12:19:58 akoww-a9mega ollama[349347]: net/http/server.go:690 +0x37 fp=0xc00098bfc8 sp=0xc00098bf80 pc=0x55a30d118c97 Feb 26 12:19:58 akoww-a9mega ollama[349347]: net/http.(*connReader).startBackgroundRead.gowrap2() Feb 26 12:19:58 akoww-a9mega ollama[349347]: net/http/server.go:686 +0x25 fp=0xc00098bfe0 sp=0xc00098bfc8 pc=0x55a30d118bc5 Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime.goexit({}) Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime/asm_amd64.s:1700 +0x1 fp=0xc00098bfe8 sp=0xc00098bfe0 pc=0x55a30ce27e61 Feb 26 12:19:58 akoww-a9mega ollama[349347]: created by net/http.(*connReader).startBackgroundRead in goroutine 55 Feb 26 12:19:58 akoww-a9mega ollama[349347]: net/http/server.go:686 +0xb6 Feb 26 12:19:58 akoww-a9mega ollama[349347]: Feb 26 12:19:58 akoww-a9mega ollama[349347]: goroutine 971 gp=0xc0004cd880 m=nil [chan receive]: Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime.gopark(0x30?, 0x55a30e5f0d80?, 0x1?, 0xc?, 0xc000ad2b20?) Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime/proc.go:435 +0xce fp=0xc000ad2ad8 sp=0xc000ad2ab8 pc=0x55a30ce1feee Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime.chanrecv(0xc001aff5e0, 0x0, 0x1) Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime/chan.go:664 +0x445 fp=0xc000ad2b50 sp=0xc000ad2ad8 pc=0x55a30cdbc085 Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime.chanrecv1(0x55a30e117342?, 0x2c?) Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime/chan.go:506 +0x12 fp=0xc000ad2b78 sp=0xc000ad2b50 pc=0x55a30cdbbc12 Feb 26 12:19:58 akoww-a9mega ollama[349347]: github.com/ollama/ollama/runner/ollamarunner.(*Server).computeBatch(0xc0002af0e0, {0x1, {0x55a30e6c19d0, 0xc000dbb2c0}, {0x55a30e6cee30, 0xc0024823c0}, {0xc000db0358, 0x1, 0x1}, {{0x55a30e6cee30, ...}, ...}, ...}) Feb 26 12:19:58 akoww-a9mega ollama[349347]: github.com/ollama/ollama/runner/ollamarunner/runner.go:645 +0x185 fp=0xc000ad2ef0 sp=0xc000ad2b78 pc=0x55a30d440ba5 Feb 26 12:19:58 akoww-a9mega ollama[349347]: github.com/ollama/ollama/runner/ollamarunner.(*Server).run.gowrap1() Feb 26 12:19:58 akoww-a9mega ollama[349347]: github.com/ollama/ollama/runner/ollamarunner/runner.go:459 +0x58 fp=0xc000ad2fe0 sp=0xc000ad2ef0 pc=0x55a30d43ef78 Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime.goexit({}) Feb 26 12:19:58 akoww-a9mega ollama[349347]: runtime/asm_amd64.s:1700 +0x1 fp=0xc000ad2fe8 sp=0xc000ad2fe0 pc=0x55a30ce27e61 Feb 26 12:19:58 akoww-a9mega ollama[349347]: created by github.com/ollama/ollama/runner/ollamarunner.(*Server).run in goroutine 54 Feb 26 12:19:58 akoww-a9mega ollama[349347]: github.com/ollama/ollama/runner/ollamarunner/runner.go:459 +0x2cd Feb 26 12:19:58 akoww-a9mega ollama[349347]: Feb 26 12:19:58 akoww-a9mega ollama[349347]: rax 0xb35900 Feb 26 12:19:58 akoww-a9mega ollama[349347]: rbx 0x0 Feb 26 12:19:58 akoww-a9mega ollama[349347]: rcx 0x0 Feb 26 12:19:58 akoww-a9mega ollama[349347]: rdx 0x7fe7bca41f00 Feb 26 12:19:58 akoww-a9mega ollama[349347]: rdi 0x200 Feb 26 12:19:58 akoww-a9mega ollama[349347]: rsi 0x80 Feb 26 12:19:58 akoww-a9mega ollama[349347]: rbp 0x7fe7c5b1c750 Feb 26 12:19:58 akoww-a9mega ollama[349347]: rsp 0x7fe7c5b1c680 Feb 26 12:19:58 akoww-a9mega ollama[349347]: r8 0x7fe7bca41f04 Feb 26 12:19:58 akoww-a9mega ollama[349347]: r9 0x80 Feb 26 12:19:58 akoww-a9mega ollama[349347]: r10 0x7f Feb 26 12:19:58 akoww-a9mega ollama[349347]: r11 0x7f Feb 26 12:19:58 akoww-a9mega ollama[349347]: r12 0x7fe7bca41f00 Feb 26 12:19:58 akoww-a9mega ollama[349347]: r13 0xb35900 Feb 26 12:19:58 akoww-a9mega ollama[349347]: r14 0x0 Feb 26 12:19:58 akoww-a9mega ollama[349347]: r15 0x40 Feb 26 12:19:58 akoww-a9mega ollama[349347]: rip 0x7fe7c5e3957f Feb 26 12:19:58 akoww-a9mega ollama[349347]: rflags 0x10246 Feb 26 12:19:58 akoww-a9mega ollama[349347]: cs 0x33 Feb 26 12:19:58 akoww-a9mega ollama[349347]: fs 0x0 Feb 26 12:19:58 akoww-a9mega ollama[349347]: gs 0x0 Feb 26 12:19:58 akoww-a9mega ollama[349347]: time=2026-02-26T11:19:58.176Z level=ERROR source=server.go:1610 msg="post predict" error="Post \"http://127.0.0.1:42155/completion\": EOF" Feb 26 12:19:58 akoww-a9mega ollama[349347]: [GIN] 2026/02/26 - 11:19:58 | 500 | 123.372292ms | 10.88.0.1 | POST "/api/chat" ``` ### OS Linux ### GPU AMD ### CPU AMD ### Ollama version 0.17.1
GiteaMirror added the bug label 2026-04-12 22:16:41 -05:00
GiteaMirror commented 2026-04-12 22:16:42 -05:00
Author
Owner
Copy Link

@Trossaloss commented on GitHub (Feb 26, 2026):

I'm having the same issue here. I'm using Vulkan on a Strix Halo and using .17.1. Same crash logs. This issue impacts Qwen3-Coder-Next, as well as any Qwen3.5 models.

<!-- gh-comment-id:3966810391 --> @Trossaloss commented on GitHub (Feb 26, 2026): I'm having the same issue here. I'm using Vulkan on a Strix Halo and using .17.1. Same crash logs. This issue impacts Qwen3-Coder-Next, as well as any Qwen3.5 models.
GiteaMirror commented 2026-04-12 22:16:43 -05:00
Author
Owner
Copy Link

@waltercool commented on GitHub (Feb 27, 2026):

Still happens with ollama 0.17.4

<!-- gh-comment-id:3974975401 --> @waltercool commented on GitHub (Feb 27, 2026): Still happens with ollama 0.17.4
GiteaMirror commented 2026-04-12 22:16:43 -05:00
Author
Owner
Copy Link

@waltercool commented on GitHub (Feb 27, 2026):

My personal recommendation for now is to use 0.16.3, as you can still use qwen3-code-next.

<!-- gh-comment-id:3975042953 --> @waltercool commented on GitHub (Feb 27, 2026): My personal recommendation for now is to use 0.16.3, as you can still use qwen3-code-next.
GiteaMirror commented 2026-04-12 22:16:44 -05:00
Author
Owner
Copy Link

@mpohoril commented on GitHub (Feb 27, 2026):

  • It worked well two or three days ago. But started failing today (27.02.2026) after upgrading to latest Ollama.
  • It started crashing after update, I believe, to 0.17.4. But I also checked on 0.17.3. Same behavior WHEN(!) trying to run it with GPU (Vulkan, AMD Radeon 780M), CPU AMD Ryzen 7, RAM 64GB.
  • It works excellently without GPU (OLLAMA_VULKAN=0) on the latest Ollama 0.17.4

Agree with @waltercool. 0.16.3 is working on GPU. But slightly slower, than 0.17.4 on CPU (8.31 vs 8.64 tokens/s)

<!-- gh-comment-id:3975563870 --> @mpohoril commented on GitHub (Feb 27, 2026): - It worked well two or three days ago. But started failing today (27.02.2026) after upgrading to latest Ollama. - It started crashing after update, I believe, to 0.17.4. But I also checked on 0.17.3. Same behavior WHEN(!) trying to run it with GPU (Vulkan, AMD Radeon 780M), CPU AMD Ryzen 7, RAM 64GB. - **It works excellently without GPU** (OLLAMA_VULKAN=0) on the latest Ollama 0.17.4 Agree with @waltercool. 0.16.3 is working on GPU. But slightly slower, than 0.17.4 on CPU (8.31 vs 8.64 tokens/s)
GiteaMirror commented 2026-04-12 22:16:45 -05:00
Author
Owner
Copy Link

@giuseppes-space commented on GitHub (Feb 28, 2026):

I'm using a Nvidia GPU and seeing the same issue with qwen3-coder-next:latest crashing on versions of 0.17. I tried 2 different versions of 0.17 and they both crashed with the same error. I had to downgrade to version 0.16.0 to get coder-next working again. Here's the log file error:


[GIN] 2026/02/28 - 14:50:47 | 200 |      9.9858ms |       127.0.0.1 | HEAD     "/"
[GIN] 2026/02/28 - 14:50:47 | 200 |      3.1196ms |       127.0.0.1 | GET      "/api/ps"
time=2026-02-28T14:50:54.041-06:00 level=INFO source=server.go:1388 msg="llama runner started in 17.70 seconds"
CUDA error: invalid argument
  current device: 0, in function ggml_cuda_cpy at C:\a\ollama\ollama\ml\backend\ggml\ggml\src\ggml-cuda\cpy.cu:438
  cudaMemcpyAsyncReserve(src1_ddc, src0_ddc, ggml_nbytes(src0), cudaMemcpyDeviceToDevice, main_stream)
C:\a\ollama\ollama\ml\backend\ggml\ggml\src\ggml-cuda\ggml-cuda.cu:94: CUDA error
time=2026-02-28T14:50:54.401-06:00 level=ERROR source=server.go:1610 msg="post predict" error="Post \"http://127.0.0.1:57077/completion\": read tcp 127.0.0.1:57082->127.0.0.1:57077: wsarecv: An existing connection was forcibly closed by the remote host."
<!-- gh-comment-id:3977853732 --> @giuseppes-space commented on GitHub (Feb 28, 2026): I'm using a Nvidia GPU and seeing the same issue with qwen3-coder-next:latest crashing on versions of 0.17. I tried 2 different versions of 0.17 and they both crashed with the same error. I had to downgrade to version 0.16.0 to get coder-next working again. Here's the log file error: ``` [GIN] 2026/02/28 - 14:50:47 | 200 | 9.9858ms | 127.0.0.1 | HEAD "/" [GIN] 2026/02/28 - 14:50:47 | 200 | 3.1196ms | 127.0.0.1 | GET "/api/ps" time=2026-02-28T14:50:54.041-06:00 level=INFO source=server.go:1388 msg="llama runner started in 17.70 seconds" CUDA error: invalid argument current device: 0, in function ggml_cuda_cpy at C:\a\ollama\ollama\ml\backend\ggml\ggml\src\ggml-cuda\cpy.cu:438 cudaMemcpyAsyncReserve(src1_ddc, src0_ddc, ggml_nbytes(src0), cudaMemcpyDeviceToDevice, main_stream) C:\a\ollama\ollama\ml\backend\ggml\ggml\src\ggml-cuda\ggml-cuda.cu:94: CUDA error time=2026-02-28T14:50:54.401-06:00 level=ERROR source=server.go:1610 msg="post predict" error="Post \"http://127.0.0.1:57077/completion\": read tcp 127.0.0.1:57082->127.0.0.1:57077: wsarecv: An existing connection was forcibly closed by the remote host." ```
GiteaMirror commented 2026-04-12 22:16:46 -05:00
Author
Owner
Copy Link

@Shotsfiredbyj commented on GitHub (Mar 10, 2026):

Confirming on Ollama 0.17.1 with Vulkan on Strix Halo (128GB, Radeon 8060S, gfx1151).

qwen3-coder-next loads all 49 layers to GPU successfully but crashes with SIGSEGV during inference. Tested with OLLAMA_KV_CACHE_TYPE=q8_0 and OLLAMA_FLASH_ATTENTION=1, as well as reduced context (num_ctx 4096) — same crash regardless.

Notably, qwen3-next:80b-a3b-instruct-q4_K_M (same architecture, different fine-tune) works perfectly on the same setup with full 262K context.

OS: Nobara 43 (Fedora-based), Kernel 6.19.5

<!-- gh-comment-id:4033781751 --> @Shotsfiredbyj commented on GitHub (Mar 10, 2026): Confirming on Ollama 0.17.1 with Vulkan on Strix Halo (128GB, Radeon 8060S, gfx1151). `qwen3-coder-next` loads all 49 layers to GPU successfully but crashes with SIGSEGV during inference. Tested with `OLLAMA_KV_CACHE_TYPE=q8_0` and `OLLAMA_FLASH_ATTENTION=1`, as well as reduced context (`num_ctx` 4096) — same crash regardless. Notably, `qwen3-next:80b-a3b-instruct-q4_K_M` (same architecture, different fine-tune) works perfectly on the same setup with full 262K context. OS: Nobara 43 (Fedora-based), Kernel 6.19.5
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels
amd
api
app
bug
build
cli
cloud
compatibility
context-length
create
docker
documentation
embeddings
feature request
feedback wanted
good first issue
gpt-oss
gpu
harmony
help wanted
image
install
intel
js
launch
linux
macos
memory
mlx
model
needs more info
networking
nvidia
ollama.com
performance
pull-request
Mirrored from GitHub Pull Request
 python
question
registry
rendering
thinking
tools
top
vulkan
windows
wsl
No Label bug
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
GiteaMirror ninjasurge
No Assignees
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: github-starred/ollama#9375
Delete Branch "%!s()"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?