github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-08 00:51:34 -05:00
Code Issues 14.5k Packages Projects Releases 120 Wiki Activity

[GH-ISSUE #14321] Recently, multimodal image models have tight imagesize restriction in the API but not on the command line #55832

New Issue
Closed
opened 2026-04-29 09:47:11 -05:00 by GiteaMirror · 9 comments
GiteaMirror commented 2026-04-29 09:47:11 -05:00
Owner
Copy Link

Originally created by @daboe01 on GitHub (Feb 19, 2026).
Original GitHub issue: https://github.com/ollama/ollama/issues/14321

What is the issue?

e.g. olm-ocr gives gibberish output on an "easy" image using the rest API (CLI works fine).
reducing image size dives good output on API (and CLI).
this bug has been introduced recently.

Relevant log output

OS

No response

GPU

No response

CPU

No response

Ollama version

No response

Originally created by @daboe01 on GitHub (Feb 19, 2026). Original GitHub issue: https://github.com/ollama/ollama/issues/14321 ### What is the issue? e.g. olm-ocr gives gibberish output on an "easy" image using the rest API (CLI works fine). reducing image size dives good output on API (and CLI). this bug has been introduced recently. ### Relevant log output ```shell ``` ### OS _No response_ ### GPU _No response_ ### CPU _No response_ ### Ollama version _No response_
GiteaMirror added the bug label 2026-04-29 09:47:11 -05:00
GiteaMirror commented 2026-04-29 09:47:13 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Feb 19, 2026):

Server logs may help in debugging.

<!-- gh-comment-id:3925777635 --> @rick-github commented on GitHub (Feb 19, 2026): [Server logs]( https://github.com/ollama/ollama/blob/main/docs/troubleshooting.mdx) may help in debugging.
GiteaMirror commented 2026-04-29 09:47:13 -05:00
Author
Owner
Copy Link

@daboe01 commented on GitHub (Feb 19, 2026):

macosx@AUG-13582 VIPSPatchbay % tail -f ~/.ollama/logs/server.log
time=2026-02-19T13:38:35.528+01:00 level=ERROR source=routes.go:1878 msg="Post "https://ollama.com/api/me?ts=1771504715": read tcp 10.210.21.201:60931->34.36.133.15:443: read: connection reset by peer"
[GIN] 2026/02/19 - 13:38:35 | 200 | 24.684709ms | 127.0.0.1 | POST "/api/me"
time=2026-02-19T13:38:37.557+01:00 level=ERROR source=routes.go:1878 msg="Post "https://ollama.com/api/me?ts=1771504717": read tcp 10.210.21.201:60937->34.36.133.15:443: read: connection reset by peer"
[GIN] 2026/02/19 - 13:38:37 | 200 | 21.709417ms | 127.0.0.1 | POST "/api/me"
time=2026-02-19T13:38:39.585+01:00 level=ERROR source=routes.go:1878 msg="Post "https://ollama.com/api/me?ts=1771504719": read tcp 10.210.21.201:60943->34.36.133.15:443: read: connection reset by peer"
[GIN] 2026/02/19 - 13:38:39 | 200 | 19.717125ms | 127.0.0.1 | POST "/api/me"
time=2026-02-19T13:38:41.618+01:00 level=ERROR source=routes.go:1878 msg="Post "https://ollama.com/api/me?ts=1771504721": read tcp 10.210.21.201:60949->34.36.133.15:443: read: connection reset by peer"
[GIN] 2026/02/19 - 13:38:41 | 200 | 25.05025ms | 127.0.0.1 | POST "/api/me"
time=2026-02-19T13:38:43.646+01:00 level=ERROR source=routes.go:1878 msg="Post "https://ollama.com/api/me?ts=1771504723": read tcp 10.210.21.201:60955->34.36.133.15:443: read: connection reset by peer"
[GIN] 2026/02/19 - 13:38:43 | 200 | 20.301583ms | 127.0.0.1 | POST "/api/me"
time=2026-02-19T13:42:51.889+01:00 level=INFO source=server.go:247 msg="enabling flash attention"
time=2026-02-19T13:42:51.889+01:00 level=INFO source=server.go:431 msg="starting runner" cmd="/Applications/Ollama.app/Contents/Resources/ollama runner --ollama-engine --model /Users/macosx/.ollama/models/blobs/sha256-65493e1f85b9ea4ba3ed793515fde13cbdbea7d74ad2c662b566b146eab0081e --port 61582"
time=2026-02-19T13:42:51.891+01:00 level=INFO source=sched.go:466 msg="system memory" total="64.0 GiB" free="58.8 GiB" free_swap="0 B"
time=2026-02-19T13:42:51.891+01:00 level=INFO source=sched.go:473 msg="gpu memory" id=0 library=Metal available="47.5 GiB" free="48.0 GiB" minimum="512.0 MiB" overhead="0 B"
time=2026-02-19T13:42:51.891+01:00 level=INFO source=server.go:757 msg="loading model" "model layers"=17 requested=-1
time=2026-02-19T13:42:51.918+01:00 level=INFO source=runner.go:1411 msg="starting ollama engine"
time=2026-02-19T13:42:51.919+01:00 level=INFO source=runner.go:1446 msg="Server listening on 127.0.0.1:61582"
time=2026-02-19T13:42:51.924+01:00 level=INFO source=runner.go:1284 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Enabled KvSize:2048 KvCacheType: NumThreads:10 GPULayers:17[ID:0 Layers:17(0..16)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2026-02-19T13:42:51.930+01:00 level=INFO source=ggml.go:136 msg="" architecture=glmocr file_type=F16 name="" description="" num_tensors=527 num_key_values=47
ggml_metal_device_init: tensor API disabled for pre-M5 and pre-A19 devices
ggml_metal_library_init: using embedded metal library
ggml_metal_library_init: loaded in 0.006 sec
ggml_metal_rsets_init: creating a residency set collection (keep_alive = 180 s)
ggml_metal_device_init: GPU name: Apple M4 Pro
ggml_metal_device_init: GPU family: MTLGPUFamilyApple9 (1009)
ggml_metal_device_init: GPU family: MTLGPUFamilyCommon3 (3003)
ggml_metal_device_init: GPU family: MTLGPUFamilyMetal3 (5001)
ggml_metal_device_init: simdgroup reduction = true
ggml_metal_device_init: simdgroup matrix mul. = true
ggml_metal_device_init: has unified memory = true
ggml_metal_device_init: has bfloat = true
ggml_metal_device_init: has tensor = false
ggml_metal_device_init: use residency sets = true
ggml_metal_device_init: use shared buffers = true
ggml_metal_device_init: recommendedMaxWorkingSetSize = 51539.61 MB
time=2026-02-19T13:42:51.932+01:00 level=INFO source=ggml.go:104 msg=system Metal.0.EMBED_LIBRARY=1 CPU.0.NEON=1 CPU.0.ARM_FMA=1 CPU.0.FP16_VA=1 CPU.0.DOTPROD=1 CPU.0.LLAMAFILE=1 CPU.0.ACCELERATE=1 compiler=cgo(clang)
ggml_metal_init: allocating
ggml_metal_init: picking default device: Apple M4 Pro
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
time=2026-02-19T13:42:52.121+01:00 level=INFO source=runner.go:1284 msg=load request="{Operation:alloc LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Enabled KvSize:2048 KvCacheType: NumThreads:10 GPULayers:17[ID:0 Layers:17(0..16)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2026-02-19T13:42:52.306+01:00 level=INFO source=runner.go:1284 msg=load request="{Operation:commit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Enabled KvSize:2048 KvCacheType: NumThreads:10 GPULayers:17[ID:0 Layers:17(0..16)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2026-02-19T13:42:52.306+01:00 level=INFO source=ggml.go:482 msg="offloading 16 repeating layers to GPU"
time=2026-02-19T13:42:52.306+01:00 level=INFO source=ggml.go:489 msg="offloading output layer to GPU"
time=2026-02-19T13:42:52.306+01:00 level=INFO source=ggml.go:494 msg="offloaded 17/17 layers to GPU"
time=2026-02-19T13:42:52.306+01:00 level=INFO source=device.go:240 msg="model weights" device=Metal size="1.9 GiB"
time=2026-02-19T13:42:52.306+01:00 level=INFO source=device.go:245 msg="model weights" device=CPU size="174.0 MiB"
time=2026-02-19T13:42:52.306+01:00 level=INFO source=device.go:251 msg="kv cache" device=Metal size="128.0 MiB"
time=2026-02-19T13:42:52.306+01:00 level=INFO source=device.go:262 msg="compute graph" device=Metal size="1.2 GiB"
time=2026-02-19T13:42:52.306+01:00 level=INFO source=device.go:267 msg="compute graph" device=CPU size="31.2 MiB"
time=2026-02-19T13:42:52.306+01:00 level=INFO source=device.go:272 msg="total memory" size="3.4 GiB"
time=2026-02-19T13:42:52.306+01:00 level=INFO source=sched.go:540 msg="loaded runners" count=1
time=2026-02-19T13:42:52.306+01:00 level=INFO source=server.go:1350 msg="waiting for llama runner to start responding"
time=2026-02-19T13:42:52.306+01:00 level=INFO source=server.go:1384 msg="waiting for server to become available" status="llm server loading model"
time=2026-02-19T13:42:53.059+01:00 level=INFO source=server.go:1388 msg="llama runner started in 1.17 seconds"
time=2026-02-19T13:42:53.203+01:00 level=WARN source=runner.go:187 msg="truncating input prompt" limit=2048 prompt=3562 keep=4 new=11
[GIN] 2026/02/19 - 13:42:53 | 200 | 1.585488042s | 10.210.21.37 | POST "/api/generate"

<!-- gh-comment-id:3927059032 --> @daboe01 commented on GitHub (Feb 19, 2026): macosx@AUG-13582 VIPSPatchbay % tail -f ~/.ollama/logs/server.log time=2026-02-19T13:38:35.528+01:00 level=ERROR source=routes.go:1878 msg="Post \"https://ollama.com/api/me?ts=1771504715\": read tcp 10.210.21.201:60931->34.36.133.15:443: read: connection reset by peer" [GIN] 2026/02/19 - 13:38:35 | 200 | 24.684709ms | 127.0.0.1 | POST "/api/me" time=2026-02-19T13:38:37.557+01:00 level=ERROR source=routes.go:1878 msg="Post \"https://ollama.com/api/me?ts=1771504717\": read tcp 10.210.21.201:60937->34.36.133.15:443: read: connection reset by peer" [GIN] 2026/02/19 - 13:38:37 | 200 | 21.709417ms | 127.0.0.1 | POST "/api/me" time=2026-02-19T13:38:39.585+01:00 level=ERROR source=routes.go:1878 msg="Post \"https://ollama.com/api/me?ts=1771504719\": read tcp 10.210.21.201:60943->34.36.133.15:443: read: connection reset by peer" [GIN] 2026/02/19 - 13:38:39 | 200 | 19.717125ms | 127.0.0.1 | POST "/api/me" time=2026-02-19T13:38:41.618+01:00 level=ERROR source=routes.go:1878 msg="Post \"https://ollama.com/api/me?ts=1771504721\": read tcp 10.210.21.201:60949->34.36.133.15:443: read: connection reset by peer" [GIN] 2026/02/19 - 13:38:41 | 200 | 25.05025ms | 127.0.0.1 | POST "/api/me" time=2026-02-19T13:38:43.646+01:00 level=ERROR source=routes.go:1878 msg="Post \"https://ollama.com/api/me?ts=1771504723\": read tcp 10.210.21.201:60955->34.36.133.15:443: read: connection reset by peer" [GIN] 2026/02/19 - 13:38:43 | 200 | 20.301583ms | 127.0.0.1 | POST "/api/me" time=2026-02-19T13:42:51.889+01:00 level=INFO source=server.go:247 msg="enabling flash attention" time=2026-02-19T13:42:51.889+01:00 level=INFO source=server.go:431 msg="starting runner" cmd="/Applications/Ollama.app/Contents/Resources/ollama runner --ollama-engine --model /Users/macosx/.ollama/models/blobs/sha256-65493e1f85b9ea4ba3ed793515fde13cbdbea7d74ad2c662b566b146eab0081e --port 61582" time=2026-02-19T13:42:51.891+01:00 level=INFO source=sched.go:466 msg="system memory" total="64.0 GiB" free="58.8 GiB" free_swap="0 B" time=2026-02-19T13:42:51.891+01:00 level=INFO source=sched.go:473 msg="gpu memory" id=0 library=Metal available="47.5 GiB" free="48.0 GiB" minimum="512.0 MiB" overhead="0 B" time=2026-02-19T13:42:51.891+01:00 level=INFO source=server.go:757 msg="loading model" "model layers"=17 requested=-1 time=2026-02-19T13:42:51.918+01:00 level=INFO source=runner.go:1411 msg="starting ollama engine" time=2026-02-19T13:42:51.919+01:00 level=INFO source=runner.go:1446 msg="Server listening on 127.0.0.1:61582" time=2026-02-19T13:42:51.924+01:00 level=INFO source=runner.go:1284 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Enabled KvSize:2048 KvCacheType: NumThreads:10 GPULayers:17[ID:0 Layers:17(0..16)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2026-02-19T13:42:51.930+01:00 level=INFO source=ggml.go:136 msg="" architecture=glmocr file_type=F16 name="" description="" num_tensors=527 num_key_values=47 ggml_metal_device_init: tensor API disabled for pre-M5 and pre-A19 devices ggml_metal_library_init: using embedded metal library ggml_metal_library_init: loaded in 0.006 sec ggml_metal_rsets_init: creating a residency set collection (keep_alive = 180 s) ggml_metal_device_init: GPU name: Apple M4 Pro ggml_metal_device_init: GPU family: MTLGPUFamilyApple9 (1009) ggml_metal_device_init: GPU family: MTLGPUFamilyCommon3 (3003) ggml_metal_device_init: GPU family: MTLGPUFamilyMetal3 (5001) ggml_metal_device_init: simdgroup reduction = true ggml_metal_device_init: simdgroup matrix mul. = true ggml_metal_device_init: has unified memory = true ggml_metal_device_init: has bfloat = true ggml_metal_device_init: has tensor = false ggml_metal_device_init: use residency sets = true ggml_metal_device_init: use shared buffers = true ggml_metal_device_init: recommendedMaxWorkingSetSize = 51539.61 MB time=2026-02-19T13:42:51.932+01:00 level=INFO source=ggml.go:104 msg=system Metal.0.EMBED_LIBRARY=1 CPU.0.NEON=1 CPU.0.ARM_FMA=1 CPU.0.FP16_VA=1 CPU.0.DOTPROD=1 CPU.0.LLAMAFILE=1 CPU.0.ACCELERATE=1 compiler=cgo(clang) ggml_metal_init: allocating ggml_metal_init: picking default device: Apple M4 Pro ggml_metal_init: use fusion = true ggml_metal_init: use concurrency = true ggml_metal_init: use graph optimize = true time=2026-02-19T13:42:52.121+01:00 level=INFO source=runner.go:1284 msg=load request="{Operation:alloc LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Enabled KvSize:2048 KvCacheType: NumThreads:10 GPULayers:17[ID:0 Layers:17(0..16)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2026-02-19T13:42:52.306+01:00 level=INFO source=runner.go:1284 msg=load request="{Operation:commit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Enabled KvSize:2048 KvCacheType: NumThreads:10 GPULayers:17[ID:0 Layers:17(0..16)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2026-02-19T13:42:52.306+01:00 level=INFO source=ggml.go:482 msg="offloading 16 repeating layers to GPU" time=2026-02-19T13:42:52.306+01:00 level=INFO source=ggml.go:489 msg="offloading output layer to GPU" time=2026-02-19T13:42:52.306+01:00 level=INFO source=ggml.go:494 msg="offloaded 17/17 layers to GPU" time=2026-02-19T13:42:52.306+01:00 level=INFO source=device.go:240 msg="model weights" device=Metal size="1.9 GiB" time=2026-02-19T13:42:52.306+01:00 level=INFO source=device.go:245 msg="model weights" device=CPU size="174.0 MiB" time=2026-02-19T13:42:52.306+01:00 level=INFO source=device.go:251 msg="kv cache" device=Metal size="128.0 MiB" time=2026-02-19T13:42:52.306+01:00 level=INFO source=device.go:262 msg="compute graph" device=Metal size="1.2 GiB" time=2026-02-19T13:42:52.306+01:00 level=INFO source=device.go:267 msg="compute graph" device=CPU size="31.2 MiB" time=2026-02-19T13:42:52.306+01:00 level=INFO source=device.go:272 msg="total memory" size="3.4 GiB" time=2026-02-19T13:42:52.306+01:00 level=INFO source=sched.go:540 msg="loaded runners" count=1 time=2026-02-19T13:42:52.306+01:00 level=INFO source=server.go:1350 msg="waiting for llama runner to start responding" time=2026-02-19T13:42:52.306+01:00 level=INFO source=server.go:1384 msg="waiting for server to become available" status="llm server loading model" time=2026-02-19T13:42:53.059+01:00 level=INFO source=server.go:1388 msg="llama runner started in 1.17 seconds" time=2026-02-19T13:42:53.203+01:00 level=WARN source=runner.go:187 msg="truncating input prompt" limit=2048 prompt=3562 keep=4 new=11 [GIN] 2026/02/19 - 13:42:53 | 200 | 1.585488042s | 10.210.21.37 | POST "/api/generate"
GiteaMirror commented 2026-04-29 09:47:14 -05:00
Author
Owner
Copy Link

@daboe01 commented on GitHub (Feb 19, 2026):

i suspect it is the truncating input prompt thing

<!-- gh-comment-id:3927060834 --> @daboe01 commented on GitHub (Feb 19, 2026): i suspect it is the **truncating input prompt** thing
GiteaMirror commented 2026-04-29 09:47:14 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Feb 19, 2026):

i suspect it is the truncating input prompt thing

Probably. The model has been loaded with a context length of 2048. Since this is not a complete log, it's unclear if this is a server config issue or a client issue. Since the CLI works fine, it's likely a client issue. Configure the client to either not set num_ctx, or configure it to set it to a larger value.

<!-- gh-comment-id:3927244128 --> @rick-github commented on GitHub (Feb 19, 2026): > i suspect it is the truncating input prompt thing Probably. The model has been loaded with a context length of 2048. Since this is not a complete log, it's unclear if this is a server config issue or a client issue. Since the CLI works fine, it's likely a client issue. Configure the client to either not set `num_ctx`, or configure it to set it to a larger value.
GiteaMirror commented 2026-04-29 09:47:15 -05:00
Author
Owner
Copy Link

@daboe01 commented on GitHub (Feb 19, 2026):

setting it to a larger value (4000) causes a crash:

time=2026-02-19T17:00:27.755+01:00 level=INFO source=server.go:247 msg="enabling flash attention"
time=2026-02-19T17:00:27.755+01:00 level=INFO source=server.go:431 msg="starting runner" cmd="/Applications/Ollama.app/Contents/Resources/ollama runner --ollama-engine --model /Users/macosx/.ollama/models/blobs/sha256-65493e1f85b9ea4ba3ed793515fde13cbdbea7d74ad2c662b566b146eab0081e --port 58843"
time=2026-02-19T17:00:27.757+01:00 level=INFO source=sched.go:466 msg="system memory" total="64.0 GiB" free="59.0 GiB" free_swap="0 B"
time=2026-02-19T17:00:27.757+01:00 level=INFO source=sched.go:473 msg="gpu memory" id=0 library=Metal available="47.5 GiB" free="48.0 GiB" minimum="512.0 MiB" overhead="0 B"
time=2026-02-19T17:00:27.757+01:00 level=INFO source=server.go:757 msg="loading model" "model layers"=17 requested=-1
time=2026-02-19T17:00:27.786+01:00 level=INFO source=runner.go:1411 msg="starting ollama engine"
time=2026-02-19T17:00:27.786+01:00 level=INFO source=runner.go:1446 msg="Server listening on 127.0.0.1:58843"
time=2026-02-19T17:00:27.790+01:00 level=INFO source=runner.go:1284 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Enabled KvSize:4000 KvCacheType: NumThreads:10 GPULayers:17[ID:0 Layers:17(0..16)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2026-02-19T17:00:27.797+01:00 level=INFO source=ggml.go:136 msg="" architecture=glmocr file_type=F16 name="" description="" num_tensors=527 num_key_values=47
ggml_metal_device_init: tensor API disabled for pre-M5 and pre-A19 devices
ggml_metal_library_init: using embedded metal library
ggml_metal_library_init: loaded in 0.006 sec
ggml_metal_rsets_init: creating a residency set collection (keep_alive = 180 s)
ggml_metal_device_init: GPU name: Apple M4 Pro
ggml_metal_device_init: GPU family: MTLGPUFamilyApple9 (1009)
ggml_metal_device_init: GPU family: MTLGPUFamilyCommon3 (3003)
ggml_metal_device_init: GPU family: MTLGPUFamilyMetal3 (5001)
ggml_metal_device_init: simdgroup reduction = true
ggml_metal_device_init: simdgroup matrix mul. = true
ggml_metal_device_init: has unified memory = true
ggml_metal_device_init: has bfloat = true
ggml_metal_device_init: has tensor = false
ggml_metal_device_init: use residency sets = true
ggml_metal_device_init: use shared buffers = true
ggml_metal_device_init: recommendedMaxWorkingSetSize = 51539.61 MB
time=2026-02-19T17:00:27.798+01:00 level=INFO source=ggml.go:104 msg=system Metal.0.EMBED_LIBRARY=1 CPU.0.NEON=1 CPU.0.ARM_FMA=1 CPU.0.FP16_VA=1 CPU.0.DOTPROD=1 CPU.0.LLAMAFILE=1 CPU.0.ACCELERATE=1 compiler=cgo(clang)
ggml_metal_init: allocating
ggml_metal_init: picking default device: Apple M4 Pro
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
time=2026-02-19T17:00:28.020+01:00 level=INFO source=runner.go:1284 msg=load request="{Operation:alloc LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Enabled KvSize:4000 KvCacheType: NumThreads:10 GPULayers:17[ID:0 Layers:17(0..16)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2026-02-19T17:00:28.236+01:00 level=INFO source=runner.go:1284 msg=load request="{Operation:commit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Enabled KvSize:4000 KvCacheType: NumThreads:10 GPULayers:17[ID:0 Layers:17(0..16)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2026-02-19T17:00:28.236+01:00 level=INFO source=ggml.go:482 msg="offloading 16 repeating layers to GPU"
time=2026-02-19T17:00:28.236+01:00 level=INFO source=ggml.go:489 msg="offloading output layer to GPU"
time=2026-02-19T17:00:28.236+01:00 level=INFO source=ggml.go:494 msg="offloaded 17/17 layers to GPU"
time=2026-02-19T17:00:28.236+01:00 level=INFO source=device.go:240 msg="model weights" device=Metal size="1.9 GiB"
time=2026-02-19T17:00:28.236+01:00 level=INFO source=device.go:245 msg="model weights" device=CPU size="174.0 MiB"
time=2026-02-19T17:00:28.236+01:00 level=INFO source=device.go:251 msg="kv cache" device=Metal size="256.0 MiB"
time=2026-02-19T17:00:28.236+01:00 level=INFO source=device.go:262 msg="compute graph" device=Metal size="1.2 GiB"
time=2026-02-19T17:00:28.236+01:00 level=INFO source=device.go:267 msg="compute graph" device=CPU size="31.2 MiB"
time=2026-02-19T17:00:28.236+01:00 level=INFO source=device.go:272 msg="total memory" size="3.5 GiB"
time=2026-02-19T17:00:28.236+01:00 level=INFO source=sched.go:540 msg="loaded runners" count=1
time=2026-02-19T17:00:28.237+01:00 level=INFO source=server.go:1350 msg="waiting for llama runner to start responding"
time=2026-02-19T17:00:28.237+01:00 level=INFO source=server.go:1384 msg="waiting for server to become available" status="llm server loading model"
time=2026-02-19T17:00:28.488+01:00 level=INFO source=server.go:1388 msg="llama runner started in 0.73 seconds"
ggml.c:4081: GGML_ASSERT(a->ne[2] * 4 == b->ne[0]) failed
WARNING: Using native backtrace. Set GGML_BACKTRACE_LLDB for more info.
WARNING: GGML_BACKTRACE_LLDB may cause native MacOS Terminal.app to crash.
See: https://github.com/ggml-org/llama.cpp/pull/17869
0 ollama 0x000000010324e660 ggml_print_backtrace + 276
1 ollama 0x000000010324e84c ggml_abort + 156
2 ollama 0x0000000103256e88 ggml_rope + 300
3 ollama 0x00000001032570e4 ggml_rope_multi + 20
4 ollama 0x00000001031e9fa0 _cgo_78bf35ffe8be_Cfunc_ggml_rope_multi + 64
5 ollama 0x0000000102507efc ollama + 524028
SIGABRT: abort
PC=0x1925bf720 m=8 sigcode=0
signal arrived during cgo execution

goroutine 15 gp=0x14000682c40 m=8 mp=0x14000532008 [syscall]:
runtime.cgocall(0x1031e9f60, 0x14000a811c8)
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/cgocall.go:167 +0x44 fp=0x14000a81190 sp=0x14000a81150 pc=0x1024fc7f4
github.com/ollama/ollama/ml/backend/ggml._Cfunc_ggml_rope_multi(0x60000392f660, 0x40e000190, 0x40e000020, 0x0, 0x80, 0x140092defb0, 0x8, 0x20000, 0x461c4000, 0x3f800000, ...)
cgo_gotypes.go:2178 +0x34 fp=0x14000a811c0 sp=0x14000a81190 pc=0x102963914
github.com/ollama/ollama/ml/backend/ggml.(*Tensor).RoPE.func2(...)
/Users/runner/work/ollama/ollama/ml/backend/ggml/ggml.go:1550
github.com/ollama/ollama/ml/backend/ggml.(*Tensor).RoPE(0x14000b53080, {0x103a89550, 0x140011da6c0}, {0x103a966c0, 0x14000b53068}, 0x80, 0x461c4000, 0x3f800000, {0x140004284e0, 0x1, ...})
/Users/runner/work/ollama/ollama/ml/backend/ggml/ggml.go:1550 +0x4d0 fp=0x14000a81330 sp=0x14000a811c0 pc=0x102970ba0
github.com/ollama/ollama/ml/nn.RoPE({0x103a89550?, 0x140011da6c0?}, {0x103a966c0?, 0x14000b53080?}, {0x103a966c0?, 0x14000b53068?}, 0xfffff22100a81418?, 0x74?, 0x0?, {0x140004284e0?, ...})
/Users/runner/work/ollama/ollama/ml/nn/rope.go:16 +0x84 fp=0x14000a81390 sp=0x14000a81330 pc=0x1029be5c4
github.com/ollama/ollama/model/models/glmocr.(*TextModelOptions).applyMRoPE(0x14001158e10, {0x103a89550, 0x140011da6c0}, {0x103a966c0, 0x14000b53080}, {0x103a966c0, 0x14000b53068})
/Users/runner/work/ollama/ollama/model/models/glmocr/model_text.go:27 +0x108 fp=0x14000a81420 sp=0x14000a81390 pc=0x102a0cc78
github.com/ollama/ollama/model/models/glmocr.(*TextModel).Shift(...)
/Users/runner/work/ollama/ollama/model/models/glmocr/model_text.go:135
github.com/ollama/ollama/model/models/glmocr.(*TextModel).Shift-fm({0x103a89550?, 0x140011da6c0?}, 0x140011da6c0?, {0x103a966c0?, 0x14000b53080?}, {0x103a966c0?, 0x14000b53068?})
:1 +0x74 fp=0x14000a81470 sp=0x14000a81420 pc=0x102a103e4
github.com/ollama/ollama/kvcache.(*Causal).shift(0x14000209300, 0x0, 0x4, 0xfffff221)
/Users/runner/work/ollama/ollama/kvcache/causal.go:599 +0x400 fp=0x14000a815d0 sp=0x14000a81470 pc=0x10295c950
github.com/ollama/ollama/kvcache.(*Causal).Remove(0x14000209300, 0x0, 0x4, 0xde3)
/Users/runner/work/ollama/ollama/kvcache/causal.go:659 +0x1fc fp=0x14000a81640 sp=0x14000a815d0 pc=0x10295cc5c
github.com/ollama/ollama/runner/ollamarunner.(*InputCache).ShiftCacheSlot(0x140011dac40, 0x140011dac00, 0x4)
/Users/runner/work/ollama/ollama/runner/ollamarunner/cache.go:299 +0x2c8 fp=0x14000a817d0 sp=0x14000a81640 pc=0x102a4f088
github.com/ollama/ollama/runner/ollamarunner.(*Server).forwardBatch(, {0x1b8, {0x103a89550, 0x1400052c080}, {0x103a966c0, 0x1400038a8b8}, {0x1400007e070, 0x1, 0x1}, {{0x103a966c0, ...}, ...}, ...})
/Users/runner/work/ollama/ollama/runner/ollamarunner/runner.go:559 +0xa5c fp=0x14000a81b40 sp=0x14000a817d0 pc=0x102a51c0c
github.com/ollama/ollama/runner/ollamarunner.(*Server).run(0x1400023b0e0, {0x103a7d360, 0x1400061b220})
/Users/runner/work/ollama/ollama/runner/ollamarunner/runner.go:453 +0x130 fp=0x14000a81fa0 sp=0x14000a81b40 pc=0x102a50fb0
github.com/ollama/ollama/runner/ollamarunner.Execute.gowrap1()
/Users/runner/work/ollama/ollama/runner/ollamarunner/runner.go:1424 +0x30 fp=0x14000a81fd0 sp=0x14000a81fa0 pc=0x102a58d10
runtime.goexit({})
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/asm_arm64.s:1223 +0x4 fp=0x14000a81fd0 sp=0x14000a81fd0 pc=0x102508104
created by github.com/ollama/ollama/runner/ollamarunner.Execute in goroutine 1
/Users/runner/work/ollama/ollama/runner/ollamarunner/runner.go:1424 +0x448

goroutine 1 gp=0x140000021c0 m=nil [IO wait, locked to thread]:
runtime.gopark(0x0?, 0x0?, 0x0?, 0x0?, 0x0?)
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/proc.go:435 +0xc8 fp=0x14000a85710 sp=0x14000a856f0 pc=0x1024ffd18
runtime.netpollblock(0x1400011d7a8?, 0x2584470?, 0x1?)
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/netpoll.go:575 +0x158 fp=0x14000a85750 sp=0x14000a85710 pc=0x1024c5778
internal/poll.runtime_pollWait(0x12d833510, 0x72)
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/netpoll.go:351 +0xa0 fp=0x14000a85780 sp=0x14000a85750 pc=0x1024feed0
internal/poll.(*pollDesc).wait(0x14000627580?, 0x1025866d8?, 0x0)
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/internal/poll/fd_poll_runtime.go:84 +0x28 fp=0x14000a857b0 sp=0x14000a85780 pc=0x10257fc88
internal/poll.(*pollDesc).waitRead(...)
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/internal/poll/fd_poll_runtime.go:89
internal/poll.(*FD).Accept(0x14000627580)
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/internal/poll/fd_unix.go:620 +0x24c fp=0x14000a85860 sp=0x14000a857b0 pc=0x10258455c
net.(*netFD).accept(0x14000627580)
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/net/fd_unix.go:172 +0x28 fp=0x14000a85920 sp=0x14000a85860 pc=0x1025f4728
net.(*TCPListener).accept(0x1400052d4c0)
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/net/tcpsock_posix.go:159 +0x24 fp=0x14000a85970 sp=0x14000a85920 pc=0x102608f04
net.(*TCPListener).Accept(0x1400052d4c0)
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/net/tcpsock.go:380 +0x2c fp=0x14000a859b0 sp=0x14000a85970 pc=0x102607eec
net/http.(*onceCloseListener).Accept(0x140004d8480?)
:1 +0x30 fp=0x14000a859d0 sp=0x14000a859b0 pc=0x1027e3910
net/http.(*Server).Serve(0x14000209a00, {0x103a7aa80, 0x1400052d4c0})
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/net/http/server.go:3424 +0x290 fp=0x14000a85b00 sp=0x14000a859d0 pc=0x1027bd050
github.com/ollama/ollama/runner/ollamarunner.Execute({0x14000032110, 0x4, 0x4})
/Users/runner/work/ollama/ollama/runner/ollamarunner/runner.go:1447 +0x7fc fp=0x14000a85cd0 sp=0x14000a85b00 pc=0x102a58aec
github.com/ollama/ollama/runner.Execute({0x140000320f0?, 0x0?, 0x0?})
/Users/runner/work/ollama/ollama/runner/runner.go:18 +0x14c fp=0x14000a85d10 sp=0x14000a85cd0 pc=0x102acfe4c
github.com/ollama/ollama/cmd.NewCLI.func3(0x14000209700?, {0x10350c330?, 0x4?, 0x10350c334?})
/Users/runner/work/ollama/ollama/cmd/cmd.go:2262 +0x54 fp=0x14000a85d40 sp=0x14000a85d10 pc=0x103198cc4
github.com/spf13/cobra.(*Command).execute(0x140004ddb08, {0x1400061b180, 0x5, 0x5})
/Users/runner/go/pkg/mod/github.com/spf13/cobra@v1.7.0/command.go:940 +0x648 fp=0x14000a85e60 sp=0x14000a85d40 pc=0x1026635c8
github.com/spf13/cobra.(*Command).ExecuteC(0x140004bc908)
/Users/runner/go/pkg/mod/github.com/spf13/cobra@v1.7.0/command.go:1068 +0x320 fp=0x14000a85f20 sp=0x14000a85e60 pc=0x102663d10
github.com/spf13/cobra.(*Command).Execute(...)
/Users/runner/go/pkg/mod/github.com/spf13/cobra@v1.7.0/command.go:992
github.com/spf13/cobra.(*Command).ExecuteContext(...)
/Users/runner/go/pkg/mod/github.com/spf13/cobra@v1.7.0/command.go:985
main.main()
/Users/runner/work/ollama/ollama/main.go:12 +0x54 fp=0x14000a85f40 sp=0x14000a85f20 pc=0x10319ab34
runtime.main()
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/proc.go:283 +0x284 fp=0x14000a85fd0 sp=0x14000a85f40 pc=0x1024cc2e4
runtime.goexit({})
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/asm_arm64.s:1223 +0x4 fp=0x14000a85fd0 sp=0x14000a85fd0 pc=0x102508104

goroutine 2 gp=0x14000002c40 m=nil [force gc (idle)]:
runtime.gopark(0x0?, 0x0?, 0x0?, 0x0?, 0x0?)
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/proc.go:435 +0xc8 fp=0x1400007af90 sp=0x1400007af70 pc=0x1024ffd18
runtime.goparkunlock(...)
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/proc.go:441
runtime.forcegchelper()
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/proc.go:348 +0xb8 fp=0x1400007afd0 sp=0x1400007af90 pc=0x1024cc638
runtime.goexit({})
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/asm_arm64.s:1223 +0x4 fp=0x1400007afd0 sp=0x1400007afd0 pc=0x102508104
created by runtime.init.7 in goroutine 1
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/proc.go:336 +0x24

goroutine 3 gp=0x14000003180 m=nil [GC sweep wait]:
runtime.gopark(0x1?, 0x0?, 0x0?, 0x0?, 0x0?)
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/proc.go:435 +0xc8 fp=0x1400007b760 sp=0x1400007b740 pc=0x1024ffd18
runtime.goparkunlock(...)
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/proc.go:441
runtime.bgsweep(0x140000a6000)
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgcsweep.go:316 +0x108 fp=0x1400007b7b0 sp=0x1400007b760 pc=0x1024b7718
runtime.gcenable.gowrap1()
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:204 +0x28 fp=0x1400007b7d0 sp=0x1400007b7b0 pc=0x1024ab518
runtime.goexit({})
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/asm_arm64.s:1223 +0x4 fp=0x1400007b7d0 sp=0x1400007b7d0 pc=0x102508104
created by runtime.gcenable in goroutine 1
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:204 +0x6c

goroutine 4 gp=0x14000003340 m=nil [GC scavenge wait]:
runtime.gopark(0x2f3d96?, 0x297a64?, 0x0?, 0x0?, 0x0?)
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/proc.go:435 +0xc8 fp=0x14000078f60 sp=0x14000078f40 pc=0x1024ffd18
runtime.goparkunlock(...)
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/proc.go:441
runtime.(*scavengerState).park(0x1044c2320)
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgcscavenge.go:425 +0x5c fp=0x14000078f90 sp=0x14000078f60 pc=0x1024b51ac
runtime.bgscavenge(0x140000a6000)
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgcscavenge.go:658 +0xac fp=0x14000078fb0 sp=0x14000078f90 pc=0x1024b574c
runtime.gcenable.gowrap2()
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:205 +0x28 fp=0x14000078fd0 sp=0x14000078fb0 pc=0x1024ab4b8
runtime.goexit({})
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/asm_arm64.s:1223 +0x4 fp=0x14000078fd0 sp=0x14000078fd0 pc=0x102508104
created by runtime.gcenable in goroutine 1
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:205 +0xac

goroutine 5 gp=0x14000003c00 m=nil [finalizer wait]:
runtime.gopark(0x180007a5c8?, 0x10470fef0?, 0x8?, 0x41?, 0x1c0?)
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/proc.go:435 +0xc8 fp=0x1400007a590 sp=0x1400007a570 pc=0x1024ffd18
runtime.runfinq()
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mfinal.go:196 +0x108 fp=0x1400007a7d0 sp=0x1400007a590 pc=0x1024aa518
runtime.goexit({})
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/asm_arm64.s:1223 +0x4 fp=0x1400007a7d0 sp=0x1400007a7d0 pc=0x102508104
created by runtime.createfing in goroutine 1
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mfinal.go:166 +0x80

goroutine 6 gp=0x140001f4700 m=nil [chan receive]:
runtime.gopark(0x140002514a0?, 0x14000a21f20?, 0x48?, 0xc7?, 0x1025c88f8?)
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/proc.go:435 +0xc8 fp=0x1400007c6f0 sp=0x1400007c6d0 pc=0x1024ffd18
runtime.chanrecv(0x140000b4310, 0x0, 0x1)
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/chan.go:664 +0x42c fp=0x1400007c770 sp=0x1400007c6f0 pc=0x10249c88c
runtime.chanrecv1(0x0?, 0x0?)
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/chan.go:506 +0x14 fp=0x1400007c7a0 sp=0x1400007c770 pc=0x10249c424
runtime.unique_runtime_registerUniqueMapCleanup.func2(...)
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:1796
runtime.unique_runtime_registerUniqueMapCleanup.gowrap1()
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:1799 +0x3c fp=0x1400007c7d0 sp=0x1400007c7a0 pc=0x1024ae73c
runtime.goexit({})
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/asm_arm64.s:1223 +0x4 fp=0x1400007c7d0 sp=0x1400007c7d0 pc=0x102508104
created by unique.runtime_registerUniqueMapCleanup in goroutine 1
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:1794 +0x78

goroutine 7 gp=0x140001f4e00 m=nil [GC worker (idle)]:
runtime.gopark(0x0?, 0x0?, 0x0?, 0x0?, 0x0?)
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/proc.go:435 +0xc8 fp=0x1400007cf10 sp=0x1400007cef0 pc=0x1024ffd18
runtime.gcBgMarkWorker(0x140000b5730)
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:1423 +0xdc fp=0x1400007cfb0 sp=0x1400007cf10 pc=0x1024ad9ac
runtime.gcBgMarkStartWorkers.gowrap1()
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:1339 +0x28 fp=0x1400007cfd0 sp=0x1400007cfb0 pc=0x1024ad898
runtime.goexit({})
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/asm_arm64.s:1223 +0x4 fp=0x1400007cfd0 sp=0x1400007cfd0 pc=0x102508104
created by runtime.gcBgMarkStartWorkers in goroutine 1
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:1339 +0x140

goroutine 8 gp=0x140001f4fc0 m=nil [GC worker (idle)]:
runtime.gopark(0x6bb94e1a4cda?, 0x3?, 0x35?, 0x5?, 0x0?)
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/proc.go:435 +0xc8 fp=0x1400007d710 sp=0x1400007d6f0 pc=0x1024ffd18
runtime.gcBgMarkWorker(0x140000b5730)
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:1423 +0xdc fp=0x1400007d7b0 sp=0x1400007d710 pc=0x1024ad9ac
runtime.gcBgMarkStartWorkers.gowrap1()
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:1339 +0x28 fp=0x1400007d7d0 sp=0x1400007d7b0 pc=0x1024ad898
runtime.goexit({})
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/asm_arm64.s:1223 +0x4 fp=0x1400007d7d0 sp=0x1400007d7d0 pc=0x102508104
created by runtime.gcBgMarkStartWorkers in goroutine 1
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:1339 +0x140

goroutine 9 gp=0x140001f5180 m=nil [GC worker (idle)]:
runtime.gopark(0x10450e0c0?, 0x1?, 0x64?, 0x21?, 0x0?)
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/proc.go:435 +0xc8 fp=0x1400007df10 sp=0x1400007def0 pc=0x1024ffd18
runtime.gcBgMarkWorker(0x140000b5730)
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:1423 +0xdc fp=0x1400007dfb0 sp=0x1400007df10 pc=0x1024ad9ac
runtime.gcBgMarkStartWorkers.gowrap1()
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:1339 +0x28 fp=0x1400007dfd0 sp=0x1400007dfb0 pc=0x1024ad898
runtime.goexit({})
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/asm_arm64.s:1223 +0x4 fp=0x1400007dfd0 sp=0x1400007dfd0 pc=0x102508104
created by runtime.gcBgMarkStartWorkers in goroutine 1
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:1339 +0x140

goroutine 18 gp=0x14000102380 m=nil [GC worker (idle)]:
runtime.gopark(0x6bb94e133d65?, 0x3?, 0xea?, 0x64?, 0x0?)
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/proc.go:435 +0xc8 fp=0x14000076710 sp=0x140000766f0 pc=0x1024ffd18
runtime.gcBgMarkWorker(0x140000b5730)
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:1423 +0xdc fp=0x140000767b0 sp=0x14000076710 pc=0x1024ad9ac
runtime.gcBgMarkStartWorkers.gowrap1()
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:1339 +0x28 fp=0x140000767d0 sp=0x140000767b0 pc=0x1024ad898
runtime.goexit({})
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/asm_arm64.s:1223 +0x4 fp=0x140000767d0 sp=0x140000767d0 pc=0x102508104
created by runtime.gcBgMarkStartWorkers in goroutine 1
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:1339 +0x140

goroutine 34 gp=0x14000504000 m=nil [GC worker (idle)]:
runtime.gopark(0x6bb94e13efc8?, 0x3?, 0x68?, 0x49?, 0x0?)
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/proc.go:435 +0xc8 fp=0x1400050a710 sp=0x1400050a6f0 pc=0x1024ffd18
runtime.gcBgMarkWorker(0x140000b5730)
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:1423 +0xdc fp=0x1400050a7b0 sp=0x1400050a710 pc=0x1024ad9ac
runtime.gcBgMarkStartWorkers.gowrap1()
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:1339 +0x28 fp=0x1400050a7d0 sp=0x1400050a7b0 pc=0x1024ad898
runtime.goexit({})
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/asm_arm64.s:1223 +0x4 fp=0x1400050a7d0 sp=0x1400050a7d0 pc=0x102508104
created by runtime.gcBgMarkStartWorkers in goroutine 1
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:1339 +0x140

goroutine 10 gp=0x140001f5340 m=nil [GC worker (idle)]:
runtime.gopark(0x6bb94e11eab1?, 0x14000112020?, 0x1b?, 0xa?, 0x0?)
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/proc.go:435 +0xc8 fp=0x14000506710 sp=0x140005066f0 pc=0x1024ffd18
runtime.gcBgMarkWorker(0x140000b5730)
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:1423 +0xdc fp=0x140005067b0 sp=0x14000506710 pc=0x1024ad9ac
runtime.gcBgMarkStartWorkers.gowrap1()
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:1339 +0x28 fp=0x140005067d0 sp=0x140005067b0 pc=0x1024ad898
runtime.goexit({})
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/asm_arm64.s:1223 +0x4 fp=0x140005067d0 sp=0x140005067d0 pc=0x102508104
created by runtime.gcBgMarkStartWorkers in goroutine 1
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:1339 +0x140

goroutine 19 gp=0x14000102540 m=nil [GC worker (idle)]:
runtime.gopark(0x6bb50fbb4f9a?, 0x3?, 0xf4?, 0x58?, 0x0?)
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/proc.go:435 +0xc8 fp=0x14000076f10 sp=0x14000076ef0 pc=0x1024ffd18
runtime.gcBgMarkWorker(0x140000b5730)
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:1423 +0xdc fp=0x14000076fb0 sp=0x14000076f10 pc=0x1024ad9ac
runtime.gcBgMarkStartWorkers.gowrap1()
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:1339 +0x28 fp=0x14000076fd0 sp=0x14000076fb0 pc=0x1024ad898
runtime.goexit({})
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/asm_arm64.s:1223 +0x4 fp=0x14000076fd0 sp=0x14000076fd0 pc=0x102508104
created by runtime.gcBgMarkStartWorkers in goroutine 1
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:1339 +0x140

goroutine 11 gp=0x140001f5500 m=nil [GC worker (idle)]:
runtime.gopark(0x6bb94e16fe55?, 0x1?, 0x75?, 0xc5?, 0x0?)
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/proc.go:435 +0xc8 fp=0x14000506f10 sp=0x14000506ef0 pc=0x1024ffd18
runtime.gcBgMarkWorker(0x140000b5730)
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:1423 +0xdc fp=0x14000506fb0 sp=0x14000506f10 pc=0x1024ad9ac
runtime.gcBgMarkStartWorkers.gowrap1()
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:1339 +0x28 fp=0x14000506fd0 sp=0x14000506fb0 pc=0x1024ad898
runtime.goexit({})
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/asm_arm64.s:1223 +0x4 fp=0x14000506fd0 sp=0x14000506fd0 pc=0x102508104
created by runtime.gcBgMarkStartWorkers in goroutine 1
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:1339 +0x140

goroutine 12 gp=0x140001f56c0 m=nil [GC worker (idle)]:
runtime.gopark(0x10450e0c0?, 0x1?, 0xb5?, 0x4a?, 0x0?)
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/proc.go:435 +0xc8 fp=0x14000507710 sp=0x140005076f0 pc=0x1024ffd18
runtime.gcBgMarkWorker(0x140000b5730)
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:1423 +0xdc fp=0x140005077b0 sp=0x14000507710 pc=0x1024ad9ac
runtime.gcBgMarkStartWorkers.gowrap1()
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:1339 +0x28 fp=0x140005077d0 sp=0x140005077b0 pc=0x1024ad898
runtime.goexit({})
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/asm_arm64.s:1223 +0x4 fp=0x140005077d0 sp=0x140005077d0 pc=0x102508104
created by runtime.gcBgMarkStartWorkers in goroutine 1
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:1339 +0x140

goroutine 13 gp=0x140001f5880 m=nil [GC worker (idle)]:
runtime.gopark(0x6bb94e11cd12?, 0x1?, 0x57?, 0x53?, 0x0?)
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/proc.go:435 +0xc8 fp=0x14000507f10 sp=0x14000507ef0 pc=0x1024ffd18
runtime.gcBgMarkWorker(0x140000b5730)
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:1423 +0xdc fp=0x14000507fb0 sp=0x14000507f10 pc=0x1024ad9ac
runtime.gcBgMarkStartWorkers.gowrap1()
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:1339 +0x28 fp=0x14000507fd0 sp=0x14000507fb0 pc=0x1024ad898
runtime.goexit({})
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/asm_arm64.s:1223 +0x4 fp=0x14000507fd0 sp=0x14000507fd0 pc=0x102508104
created by runtime.gcBgMarkStartWorkers in goroutine 1
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:1339 +0x140

goroutine 35 gp=0x140005041c0 m=nil [GC worker (idle)]:
runtime.gopark(0x6bb94e16fb91?, 0x3?, 0xa0?, 0x78?, 0x0?)
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/proc.go:435 +0xc8 fp=0x1400050af10 sp=0x1400050aef0 pc=0x1024ffd18
runtime.gcBgMarkWorker(0x140000b5730)
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:1423 +0xdc fp=0x1400050afb0 sp=0x1400050af10 pc=0x1024ad9ac
runtime.gcBgMarkStartWorkers.gowrap1()
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:1339 +0x28 fp=0x1400050afd0 sp=0x1400050afb0 pc=0x1024ad898
runtime.goexit({})
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/asm_arm64.s:1223 +0x4 fp=0x1400050afd0 sp=0x1400050afd0 pc=0x102508104
created by runtime.gcBgMarkStartWorkers in goroutine 1
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:1339 +0x140

goroutine 14 gp=0x140001f5a40 m=nil [GC worker (idle)]:
runtime.gopark(0x6bb94e1ab3e1?, 0x1?, 0xfe?, 0xa6?, 0x0?)
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/proc.go:435 +0xc8 fp=0x14000508710 sp=0x140005086f0 pc=0x1024ffd18
runtime.gcBgMarkWorker(0x140000b5730)
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:1423 +0xdc fp=0x140005087b0 sp=0x14000508710 pc=0x1024ad9ac
runtime.gcBgMarkStartWorkers.gowrap1()
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:1339 +0x28 fp=0x140005087d0 sp=0x140005087b0 pc=0x1024ad898
runtime.goexit({})
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/asm_arm64.s:1223 +0x4 fp=0x140005087d0 sp=0x140005087d0 pc=0x102508104
created by runtime.gcBgMarkStartWorkers in goroutine 1
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:1339 +0x140

goroutine 36 gp=0x14000504380 m=nil [GC worker (idle)]:
runtime.gopark(0x6bb94e1a7684?, 0x1?, 0x7c?, 0x8?, 0x0?)
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/proc.go:435 +0xc8 fp=0x1400050b710 sp=0x1400050b6f0 pc=0x1024ffd18
runtime.gcBgMarkWorker(0x140000b5730)
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:1423 +0xdc fp=0x1400050b7b0 sp=0x1400050b710 pc=0x1024ad9ac
runtime.gcBgMarkStartWorkers.gowrap1()
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:1339 +0x28 fp=0x1400050b7d0 sp=0x1400050b7b0 pc=0x1024ad898
runtime.goexit({})
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/asm_arm64.s:1223 +0x4 fp=0x1400050b7d0 sp=0x1400050b7d0 pc=0x102508104
created by runtime.gcBgMarkStartWorkers in goroutine 1
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:1339 +0x140

goroutine 37 gp=0x14000504540 m=nil [GC worker (idle)]:
runtime.gopark(0x10450e0c0?, 0x3?, 0x84?, 0x68?, 0x0?)
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/proc.go:435 +0xc8 fp=0x1400050bf10 sp=0x1400050bef0 pc=0x1024ffd18
runtime.gcBgMarkWorker(0x140000b5730)
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:1423 +0xdc fp=0x1400050bfb0 sp=0x1400050bf10 pc=0x1024ad9ac
runtime.gcBgMarkStartWorkers.gowrap1()
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:1339 +0x28 fp=0x1400050bfd0 sp=0x1400050bfb0 pc=0x1024ad898
runtime.goexit({})
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/asm_arm64.s:1223 +0x4 fp=0x1400050bfd0 sp=0x1400050bfd0 pc=0x102508104
created by runtime.gcBgMarkStartWorkers in goroutine 1
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:1339 +0x140

goroutine 16 gp=0x14000682e00 m=nil [select]:
runtime.gopark(0x14000a83a50?, 0x2?, 0x28?, 0x6?, 0x14000a8381c?)
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/proc.go:435 +0xc8 fp=0x14000a83640 sp=0x14000a83620 pc=0x1024ffd18
runtime.selectgo(0x14000a83a50, 0x14000a83818, 0x14000626300?, 0x0, 0x1?, 0x1)
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/select.go:351 +0x6c4 fp=0x14000a83770 sp=0x14000a83640 pc=0x1024df954
github.com/ollama/ollama/runner/ollamarunner.(*Server).completion(0x1400023b0e0, {0x103a7ac60, 0x1400016c0e0}, 0x140004c2140)
/Users/runner/work/ollama/ollama/runner/ollamarunner/runner.go:956 +0xa5c fp=0x14000a83aa0 sp=0x14000a83770 pc=0x102a54a2c
github.com/ollama/ollama/runner/ollamarunner.(*Server).completion-fm({0x103a7ac60?, 0x1400016c0e0?}, 0x14000a87b28?)
:1 +0x40 fp=0x14000a83ad0 sp=0x14000a83aa0 pc=0x102a591e0
net/http.HandlerFunc.ServeHTTP(0x140005258c0?, {0x103a7ac60?, 0x1400016c0e0?}, 0x14000a87b10?)
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/net/http/server.go:2294 +0x38 fp=0x14000a83b00 sp=0x14000a83ad0 pc=0x1027b9a78
net/http.(*ServeMux).ServeHTTP(0x10?, {0x103a7ac60, 0x1400016c0e0}, 0x140004c2140)
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/net/http/server.go:2822 +0x1b4 fp=0x14000a83b50 sp=0x14000a83b00 pc=0x1027bb604
net/http.serverHandler.ServeHTTP({0x103a76f90?}, {0x103a7ac60?, 0x1400016c0e0?}, 0x1?)
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/net/http/server.go:3301 +0xbc fp=0x14000a83b80 sp=0x14000a83b50 pc=0x1027d72ec
net/http.(*conn).serve(0x140004d8480, {0x103a7d328, 0x14000234240})
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/net/http/server.go:2102 +0x52c fp=0x14000a83fa0 sp=0x14000a83b80 pc=0x1027b821c
net/http.(*Server).Serve.gowrap3()
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/net/http/server.go:3454 +0x30 fp=0x14000a83fd0 sp=0x14000a83fa0 pc=0x1027bd3e0
runtime.goexit({})
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/asm_arm64.s:1223 +0x4 fp=0x14000a83fd0 sp=0x14000a83fd0 pc=0x102508104
created by net/http.(*Server).Serve in goroutine 1
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/net/http/server.go:3454 +0x3d8

goroutine 543 gp=0x14000103dc0 m=nil [IO wait]:
runtime.gopark(0xffffffffffffffff?, 0xffffffffffffffff?, 0x23?, 0x0?, 0x102523920?)
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/proc.go:435 +0xc8 fp=0x14000187d80 sp=0x14000187d60 pc=0x1024ffd18
runtime.netpollblock(0x0?, 0x0?, 0x0?)
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/netpoll.go:575 +0x158 fp=0x14000187dc0 sp=0x14000187d80 pc=0x1024c5778
internal/poll.runtime_pollWait(0x12d8333f8, 0x72)
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/netpoll.go:351 +0xa0 fp=0x14000187df0 sp=0x14000187dc0 pc=0x1024feed0
internal/poll.(*pollDesc).wait(0x14000627600?, 0x14000234701?, 0x0)
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/internal/poll/fd_poll_runtime.go:84 +0x28 fp=0x14000187e20 sp=0x14000187df0 pc=0x10257fc88
internal/poll.(*pollDesc).waitRead(...)
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/internal/poll/fd_poll_runtime.go:89
internal/poll.(*FD).Read(0x14000627600, {0x14000234701, 0x1, 0x1})
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/internal/poll/fd_unix.go:165 +0x1fc fp=0x14000187ec0 sp=0x14000187e20 pc=0x102580f3c
net.(*netFD).Read(0x14000627600, {0x14000234701?, 0x14000187f58?, 0x1027b2c94?})
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/net/fd_posix.go:55 +0x28 fp=0x14000187f10 sp=0x14000187ec0 pc=0x1025f2cf8
net.(*conn).Read(0x14000120628, {0x14000234701?, 0xff000000ff000000?, 0xff000000ff000000?})
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/net/net.go:194 +0x34 fp=0x14000187f60 sp=0x14000187f10 pc=0x1025ffbc4
net/http.(*connReader).backgroundRead(0x140002346f0)
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/net/http/server.go:690 +0x40 fp=0x14000187fb0 sp=0x14000187f60 pc=0x1027b2b90
net/http.(*connReader).startBackgroundRead.gowrap2()
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/net/http/server.go:686 +0x28 fp=0x14000187fd0 sp=0x14000187fb0 pc=0x1027b2a78
runtime.goexit({})
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/asm_arm64.s:1223 +0x4 fp=0x14000187fd0 sp=0x14000187fd0 pc=0x102508104
created by net/http.(*connReader).startBackgroundRead in goroutine 16
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/net/http/server.go:686 +0xc4

goroutine 1529 gp=0x14000602e00 m=4 mp=0x14000081808 [syscall]:
runtime.cgocall(0x1031e8c1c, 0x140011a3a78)
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/cgocall.go:167 +0x44 fp=0x140011a3a40 sp=0x140011a3a00 pc=0x1024fc7f4
github.com/ollama/ollama/ml/backend/ggml._Cfunc_ggml_backend_sched_synchronize(0x13200aa00)
_cgo_gotypes.go:1044 +0x30 fp=0x140011a3a70 sp=0x140011a3a40 pc=0x1029606b0
github.com/ollama/ollama/ml/backend/ggml.(*Context).ComputeWithNotify.func4.1(...)
/Users/runner/work/ollama/ollama/ml/backend/ggml/ggml.go:833
github.com/ollama/ollama/ml/backend/ggml.(*Context).ComputeWithNotify.func4()
/Users/runner/work/ollama/ollama/ml/backend/ggml/ggml.go:833 +0x64 fp=0x140011a3ac0 sp=0x140011a3a70 pc=0x10296ab04
github.com/ollama/ollama/ml/backend/ggml.(*Tensor).Floats(0x1400038a8b8)
/Users/runner/work/ollama/ollama/ml/backend/ggml/ggml.go:1065 +0x94 fp=0x140011a3b50 sp=0x140011a3ac0 pc=0x10296c564
github.com/ollama/ollama/runner/ollamarunner.(*Server).computeBatch(0x1400023b0e0, {0x1b8, {0x103a89550, 0x1400052c080}, {0x103a966c0, 0x1400038a8b8}, {0x1400007e070, 0x1, 0x1}, {{0x103a966c0, ...}, ...}, ...})
/Users/runner/work/ollama/ollama/runner/ollamarunner/runner.go:723 +0x714 fp=0x140011a3ed0 sp=0x140011a3b50 pc=0x102a52d94
github.com/ollama/ollama/runner/ollamarunner.(*Server).run.gowrap1()
/Users/runner/work/ollama/ollama/runner/ollamarunner/runner.go:459 +0x5c fp=0x140011a3fd0 sp=0x140011a3ed0 pc=0x102a5117c
runtime.goexit({})
/Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/asm_arm64.s:1223 +0x4 fp=0x140011a3fd0 sp=0x140011a3fd0 pc=0x102508104
created by github.com/ollama/ollama/runner/ollamarunner.(*Server).run in goroutine 15
/Users/runner/work/ollama/ollama/runner/ollamarunner/runner.go:459 +0x22c

r0 0x0
r1 0x0
r2 0x0
r3 0x0
r4 0x192507910
r5 0x1712cdc40
r6 0x38
r7 0x0
r8 0x42e83c7e5a9e6cd8
r9 0x42e83c7f2bb29cd8
r10 0x3bb
r11 0x6
r12 0x6
r13 0x1712cd972
r14 0x102097300
r15 0x1
r16 0x148
r17 0x2045aa2c0
r18 0x0
r19 0x6
r20 0x3203
r21 0x1712cf0e0
r22 0x140092defb0
r23 0x8
r24 0x20000
r25 0x80
r26 0x1400118dde0
r27 0x828
r28 0x140005048c0
r29 0x1712ce530
lr 0x1925f7f70
sp 0x1712ce510
pc 0x1925bf720
fault 0x1925bf720
time=2026-02-19T17:00:46.955+01:00 level=ERROR source=server.go:304 msg="llama runner terminated" error="exit status 2"
[GIN] 2026/02/19 - 17:00:46 | 500 | 19.285960042s | 10.210.21.37 | POST "/api/generate"

here is the startup stuff from my log:

time=2026-02-18T08:10:12.775+01:00 level=INFO source=routes.go:1663 msg="server config" env="map[HTTPS_PROXY: HTTP_PROXY: NO_PROXY: OLLAMA_CONTEXT_LENGTH:16384 OLLAMA_DEBUG:INFO OLLAMA_EDITOR: OLLAMA_FLASH_ATTENTION:false OLLAMA_GPU_OVERHEAD:0 OLLAMA_HOST:http://0.0.0.0:11434 OLLAMA_KEEP_ALIVE:5m0s OLLAMA_KV_CACHE_TYPE: OLLAMA_LLM_LIBRARY: OLLAMA_LOAD_TIMEOUT:5m0s OLLAMA_MAX_LOADED_MODELS:0 OLLAMA_MAX_QUEUE:512 OLLAMA_MODELS:/Users/macosx/.ollama/models OLLAMA_MULTIUSER_CACHE:false OLLAMA_NEW_ENGINE:false OLLAMA_NOHISTORY:false OLLAMA_NOPRUNE:false OLLAMA_NO_CLOUD:true OLLAMA_NUM_PARALLEL:1 OLLAMA_ORIGINS:[http://localhost https://localhost http://localhost:* https://localhost:* http://127.0.0.1 https://127.0.0.1 http://127.0.0.1:* https://127.0.0.1:* http://0.0.0.0 https://0.0.0.0 http://0.0.0.0:* https://0.0.0.0:* app://* file://* tauri://* vscode-webview://* vscode-file://*] OLLAMA_REMOTES:[ollama.com] OLLAMA_SCHED_SPREAD:false http_proxy: https_proxy: no_proxy:]"

...

<!-- gh-comment-id:3928201688 --> @daboe01 commented on GitHub (Feb 19, 2026): setting it to a larger value (4000) causes a crash: time=2026-02-19T17:00:27.755+01:00 level=INFO source=server.go:247 msg="enabling flash attention" time=2026-02-19T17:00:27.755+01:00 level=INFO source=server.go:431 msg="starting runner" cmd="/Applications/Ollama.app/Contents/Resources/ollama runner --ollama-engine --model /Users/macosx/.ollama/models/blobs/sha256-65493e1f85b9ea4ba3ed793515fde13cbdbea7d74ad2c662b566b146eab0081e --port 58843" time=2026-02-19T17:00:27.757+01:00 level=INFO source=sched.go:466 msg="system memory" total="64.0 GiB" free="59.0 GiB" free_swap="0 B" time=2026-02-19T17:00:27.757+01:00 level=INFO source=sched.go:473 msg="gpu memory" id=0 library=Metal available="47.5 GiB" free="48.0 GiB" minimum="512.0 MiB" overhead="0 B" time=2026-02-19T17:00:27.757+01:00 level=INFO source=server.go:757 msg="loading model" "model layers"=17 requested=-1 time=2026-02-19T17:00:27.786+01:00 level=INFO source=runner.go:1411 msg="starting ollama engine" time=2026-02-19T17:00:27.786+01:00 level=INFO source=runner.go:1446 msg="Server listening on 127.0.0.1:58843" time=2026-02-19T17:00:27.790+01:00 level=INFO source=runner.go:1284 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Enabled KvSize:4000 KvCacheType: NumThreads:10 GPULayers:17[ID:0 Layers:17(0..16)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2026-02-19T17:00:27.797+01:00 level=INFO source=ggml.go:136 msg="" architecture=glmocr file_type=F16 name="" description="" num_tensors=527 num_key_values=47 ggml_metal_device_init: tensor API disabled for pre-M5 and pre-A19 devices ggml_metal_library_init: using embedded metal library ggml_metal_library_init: loaded in 0.006 sec ggml_metal_rsets_init: creating a residency set collection (keep_alive = 180 s) ggml_metal_device_init: GPU name: Apple M4 Pro ggml_metal_device_init: GPU family: MTLGPUFamilyApple9 (1009) ggml_metal_device_init: GPU family: MTLGPUFamilyCommon3 (3003) ggml_metal_device_init: GPU family: MTLGPUFamilyMetal3 (5001) ggml_metal_device_init: simdgroup reduction = true ggml_metal_device_init: simdgroup matrix mul. = true ggml_metal_device_init: has unified memory = true ggml_metal_device_init: has bfloat = true ggml_metal_device_init: has tensor = false ggml_metal_device_init: use residency sets = true ggml_metal_device_init: use shared buffers = true ggml_metal_device_init: recommendedMaxWorkingSetSize = 51539.61 MB time=2026-02-19T17:00:27.798+01:00 level=INFO source=ggml.go:104 msg=system Metal.0.EMBED_LIBRARY=1 CPU.0.NEON=1 CPU.0.ARM_FMA=1 CPU.0.FP16_VA=1 CPU.0.DOTPROD=1 CPU.0.LLAMAFILE=1 CPU.0.ACCELERATE=1 compiler=cgo(clang) ggml_metal_init: allocating ggml_metal_init: picking default device: Apple M4 Pro ggml_metal_init: use fusion = true ggml_metal_init: use concurrency = true ggml_metal_init: use graph optimize = true time=2026-02-19T17:00:28.020+01:00 level=INFO source=runner.go:1284 msg=load request="{Operation:alloc LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Enabled KvSize:4000 KvCacheType: NumThreads:10 GPULayers:17[ID:0 Layers:17(0..16)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2026-02-19T17:00:28.236+01:00 level=INFO source=runner.go:1284 msg=load request="{Operation:commit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Enabled KvSize:4000 KvCacheType: NumThreads:10 GPULayers:17[ID:0 Layers:17(0..16)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2026-02-19T17:00:28.236+01:00 level=INFO source=ggml.go:482 msg="offloading 16 repeating layers to GPU" time=2026-02-19T17:00:28.236+01:00 level=INFO source=ggml.go:489 msg="offloading output layer to GPU" time=2026-02-19T17:00:28.236+01:00 level=INFO source=ggml.go:494 msg="offloaded 17/17 layers to GPU" time=2026-02-19T17:00:28.236+01:00 level=INFO source=device.go:240 msg="model weights" device=Metal size="1.9 GiB" time=2026-02-19T17:00:28.236+01:00 level=INFO source=device.go:245 msg="model weights" device=CPU size="174.0 MiB" time=2026-02-19T17:00:28.236+01:00 level=INFO source=device.go:251 msg="kv cache" device=Metal size="256.0 MiB" time=2026-02-19T17:00:28.236+01:00 level=INFO source=device.go:262 msg="compute graph" device=Metal size="1.2 GiB" time=2026-02-19T17:00:28.236+01:00 level=INFO source=device.go:267 msg="compute graph" device=CPU size="31.2 MiB" time=2026-02-19T17:00:28.236+01:00 level=INFO source=device.go:272 msg="total memory" size="3.5 GiB" time=2026-02-19T17:00:28.236+01:00 level=INFO source=sched.go:540 msg="loaded runners" count=1 time=2026-02-19T17:00:28.237+01:00 level=INFO source=server.go:1350 msg="waiting for llama runner to start responding" time=2026-02-19T17:00:28.237+01:00 level=INFO source=server.go:1384 msg="waiting for server to become available" status="llm server loading model" time=2026-02-19T17:00:28.488+01:00 level=INFO source=server.go:1388 msg="llama runner started in 0.73 seconds" ggml.c:4081: GGML_ASSERT(a->ne[2] * 4 == b->ne[0]) failed WARNING: Using native backtrace. Set GGML_BACKTRACE_LLDB for more info. WARNING: GGML_BACKTRACE_LLDB may cause native MacOS Terminal.app to crash. See: https://github.com/ggml-org/llama.cpp/pull/17869 0 ollama 0x000000010324e660 ggml_print_backtrace + 276 1 ollama 0x000000010324e84c ggml_abort + 156 2 ollama 0x0000000103256e88 ggml_rope + 300 3 ollama 0x00000001032570e4 ggml_rope_multi + 20 4 ollama 0x00000001031e9fa0 _cgo_78bf35ffe8be_Cfunc_ggml_rope_multi + 64 5 ollama 0x0000000102507efc ollama + 524028 SIGABRT: abort PC=0x1925bf720 m=8 sigcode=0 signal arrived during cgo execution goroutine 15 gp=0x14000682c40 m=8 mp=0x14000532008 [syscall]: runtime.cgocall(0x1031e9f60, 0x14000a811c8) /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/cgocall.go:167 +0x44 fp=0x14000a81190 sp=0x14000a81150 pc=0x1024fc7f4 github.com/ollama/ollama/ml/backend/ggml._Cfunc_ggml_rope_multi(0x60000392f660, 0x40e000190, 0x40e000020, 0x0, 0x80, 0x140092defb0, 0x8, 0x20000, 0x461c4000, 0x3f800000, ...) _cgo_gotypes.go:2178 +0x34 fp=0x14000a811c0 sp=0x14000a81190 pc=0x102963914 github.com/ollama/ollama/ml/backend/ggml.(*Tensor).RoPE.func2(...) /Users/runner/work/ollama/ollama/ml/backend/ggml/ggml.go:1550 github.com/ollama/ollama/ml/backend/ggml.(*Tensor).RoPE(0x14000b53080, {0x103a89550, 0x140011da6c0}, {0x103a966c0, 0x14000b53068}, 0x80, 0x461c4000, 0x3f800000, {0x140004284e0, 0x1, ...}) /Users/runner/work/ollama/ollama/ml/backend/ggml/ggml.go:1550 +0x4d0 fp=0x14000a81330 sp=0x14000a811c0 pc=0x102970ba0 github.com/ollama/ollama/ml/nn.RoPE({0x103a89550?, 0x140011da6c0?}, {0x103a966c0?, 0x14000b53080?}, {0x103a966c0?, 0x14000b53068?}, 0xfffff22100a81418?, 0x74?, 0x0?, {0x140004284e0?, ...}) /Users/runner/work/ollama/ollama/ml/nn/rope.go:16 +0x84 fp=0x14000a81390 sp=0x14000a81330 pc=0x1029be5c4 github.com/ollama/ollama/model/models/glmocr.(*TextModelOptions).applyMRoPE(0x14001158e10, {0x103a89550, 0x140011da6c0}, {0x103a966c0, 0x14000b53080}, {0x103a966c0, 0x14000b53068}) /Users/runner/work/ollama/ollama/model/models/glmocr/model_text.go:27 +0x108 fp=0x14000a81420 sp=0x14000a81390 pc=0x102a0cc78 github.com/ollama/ollama/model/models/glmocr.(*TextModel).Shift(...) /Users/runner/work/ollama/ollama/model/models/glmocr/model_text.go:135 github.com/ollama/ollama/model/models/glmocr.(*TextModel).Shift-fm({0x103a89550?, 0x140011da6c0?}, 0x140011da6c0?, {0x103a966c0?, 0x14000b53080?}, {0x103a966c0?, 0x14000b53068?}) <autogenerated>:1 +0x74 fp=0x14000a81470 sp=0x14000a81420 pc=0x102a103e4 github.com/ollama/ollama/kvcache.(*Causal).shift(0x14000209300, 0x0, 0x4, 0xfffff221) /Users/runner/work/ollama/ollama/kvcache/causal.go:599 +0x400 fp=0x14000a815d0 sp=0x14000a81470 pc=0x10295c950 github.com/ollama/ollama/kvcache.(*Causal).Remove(0x14000209300, 0x0, 0x4, 0xde3) /Users/runner/work/ollama/ollama/kvcache/causal.go:659 +0x1fc fp=0x14000a81640 sp=0x14000a815d0 pc=0x10295cc5c github.com/ollama/ollama/runner/ollamarunner.(*InputCache).ShiftCacheSlot(0x140011dac40, 0x140011dac00, 0x4) /Users/runner/work/ollama/ollama/runner/ollamarunner/cache.go:299 +0x2c8 fp=0x14000a817d0 sp=0x14000a81640 pc=0x102a4f088 github.com/ollama/ollama/runner/ollamarunner.(*Server).forwardBatch(_, {0x1b8, {0x103a89550, 0x1400052c080}, {0x103a966c0, 0x1400038a8b8}, {0x1400007e070, 0x1, 0x1}, {{0x103a966c0, ...}, ...}, ...}) /Users/runner/work/ollama/ollama/runner/ollamarunner/runner.go:559 +0xa5c fp=0x14000a81b40 sp=0x14000a817d0 pc=0x102a51c0c github.com/ollama/ollama/runner/ollamarunner.(*Server).run(0x1400023b0e0, {0x103a7d360, 0x1400061b220}) /Users/runner/work/ollama/ollama/runner/ollamarunner/runner.go:453 +0x130 fp=0x14000a81fa0 sp=0x14000a81b40 pc=0x102a50fb0 github.com/ollama/ollama/runner/ollamarunner.Execute.gowrap1() /Users/runner/work/ollama/ollama/runner/ollamarunner/runner.go:1424 +0x30 fp=0x14000a81fd0 sp=0x14000a81fa0 pc=0x102a58d10 runtime.goexit({}) /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/asm_arm64.s:1223 +0x4 fp=0x14000a81fd0 sp=0x14000a81fd0 pc=0x102508104 created by github.com/ollama/ollama/runner/ollamarunner.Execute in goroutine 1 /Users/runner/work/ollama/ollama/runner/ollamarunner/runner.go:1424 +0x448 goroutine 1 gp=0x140000021c0 m=nil [IO wait, locked to thread]: runtime.gopark(0x0?, 0x0?, 0x0?, 0x0?, 0x0?) /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/proc.go:435 +0xc8 fp=0x14000a85710 sp=0x14000a856f0 pc=0x1024ffd18 runtime.netpollblock(0x1400011d7a8?, 0x2584470?, 0x1?) /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/netpoll.go:575 +0x158 fp=0x14000a85750 sp=0x14000a85710 pc=0x1024c5778 internal/poll.runtime_pollWait(0x12d833510, 0x72) /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/netpoll.go:351 +0xa0 fp=0x14000a85780 sp=0x14000a85750 pc=0x1024feed0 internal/poll.(*pollDesc).wait(0x14000627580?, 0x1025866d8?, 0x0) /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/internal/poll/fd_poll_runtime.go:84 +0x28 fp=0x14000a857b0 sp=0x14000a85780 pc=0x10257fc88 internal/poll.(*pollDesc).waitRead(...) /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/internal/poll/fd_poll_runtime.go:89 internal/poll.(*FD).Accept(0x14000627580) /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/internal/poll/fd_unix.go:620 +0x24c fp=0x14000a85860 sp=0x14000a857b0 pc=0x10258455c net.(*netFD).accept(0x14000627580) /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/net/fd_unix.go:172 +0x28 fp=0x14000a85920 sp=0x14000a85860 pc=0x1025f4728 net.(*TCPListener).accept(0x1400052d4c0) /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/net/tcpsock_posix.go:159 +0x24 fp=0x14000a85970 sp=0x14000a85920 pc=0x102608f04 net.(*TCPListener).Accept(0x1400052d4c0) /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/net/tcpsock.go:380 +0x2c fp=0x14000a859b0 sp=0x14000a85970 pc=0x102607eec net/http.(*onceCloseListener).Accept(0x140004d8480?) <autogenerated>:1 +0x30 fp=0x14000a859d0 sp=0x14000a859b0 pc=0x1027e3910 net/http.(*Server).Serve(0x14000209a00, {0x103a7aa80, 0x1400052d4c0}) /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/net/http/server.go:3424 +0x290 fp=0x14000a85b00 sp=0x14000a859d0 pc=0x1027bd050 github.com/ollama/ollama/runner/ollamarunner.Execute({0x14000032110, 0x4, 0x4}) /Users/runner/work/ollama/ollama/runner/ollamarunner/runner.go:1447 +0x7fc fp=0x14000a85cd0 sp=0x14000a85b00 pc=0x102a58aec github.com/ollama/ollama/runner.Execute({0x140000320f0?, 0x0?, 0x0?}) /Users/runner/work/ollama/ollama/runner/runner.go:18 +0x14c fp=0x14000a85d10 sp=0x14000a85cd0 pc=0x102acfe4c github.com/ollama/ollama/cmd.NewCLI.func3(0x14000209700?, {0x10350c330?, 0x4?, 0x10350c334?}) /Users/runner/work/ollama/ollama/cmd/cmd.go:2262 +0x54 fp=0x14000a85d40 sp=0x14000a85d10 pc=0x103198cc4 github.com/spf13/cobra.(*Command).execute(0x140004ddb08, {0x1400061b180, 0x5, 0x5}) /Users/runner/go/pkg/mod/github.com/spf13/cobra@v1.7.0/command.go:940 +0x648 fp=0x14000a85e60 sp=0x14000a85d40 pc=0x1026635c8 github.com/spf13/cobra.(*Command).ExecuteC(0x140004bc908) /Users/runner/go/pkg/mod/github.com/spf13/cobra@v1.7.0/command.go:1068 +0x320 fp=0x14000a85f20 sp=0x14000a85e60 pc=0x102663d10 github.com/spf13/cobra.(*Command).Execute(...) /Users/runner/go/pkg/mod/github.com/spf13/cobra@v1.7.0/command.go:992 github.com/spf13/cobra.(*Command).ExecuteContext(...) /Users/runner/go/pkg/mod/github.com/spf13/cobra@v1.7.0/command.go:985 main.main() /Users/runner/work/ollama/ollama/main.go:12 +0x54 fp=0x14000a85f40 sp=0x14000a85f20 pc=0x10319ab34 runtime.main() /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/proc.go:283 +0x284 fp=0x14000a85fd0 sp=0x14000a85f40 pc=0x1024cc2e4 runtime.goexit({}) /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/asm_arm64.s:1223 +0x4 fp=0x14000a85fd0 sp=0x14000a85fd0 pc=0x102508104 goroutine 2 gp=0x14000002c40 m=nil [force gc (idle)]: runtime.gopark(0x0?, 0x0?, 0x0?, 0x0?, 0x0?) /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/proc.go:435 +0xc8 fp=0x1400007af90 sp=0x1400007af70 pc=0x1024ffd18 runtime.goparkunlock(...) /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/proc.go:441 runtime.forcegchelper() /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/proc.go:348 +0xb8 fp=0x1400007afd0 sp=0x1400007af90 pc=0x1024cc638 runtime.goexit({}) /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/asm_arm64.s:1223 +0x4 fp=0x1400007afd0 sp=0x1400007afd0 pc=0x102508104 created by runtime.init.7 in goroutine 1 /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/proc.go:336 +0x24 goroutine 3 gp=0x14000003180 m=nil [GC sweep wait]: runtime.gopark(0x1?, 0x0?, 0x0?, 0x0?, 0x0?) /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/proc.go:435 +0xc8 fp=0x1400007b760 sp=0x1400007b740 pc=0x1024ffd18 runtime.goparkunlock(...) /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/proc.go:441 runtime.bgsweep(0x140000a6000) /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgcsweep.go:316 +0x108 fp=0x1400007b7b0 sp=0x1400007b760 pc=0x1024b7718 runtime.gcenable.gowrap1() /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:204 +0x28 fp=0x1400007b7d0 sp=0x1400007b7b0 pc=0x1024ab518 runtime.goexit({}) /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/asm_arm64.s:1223 +0x4 fp=0x1400007b7d0 sp=0x1400007b7d0 pc=0x102508104 created by runtime.gcenable in goroutine 1 /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:204 +0x6c goroutine 4 gp=0x14000003340 m=nil [GC scavenge wait]: runtime.gopark(0x2f3d96?, 0x297a64?, 0x0?, 0x0?, 0x0?) /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/proc.go:435 +0xc8 fp=0x14000078f60 sp=0x14000078f40 pc=0x1024ffd18 runtime.goparkunlock(...) /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/proc.go:441 runtime.(*scavengerState).park(0x1044c2320) /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgcscavenge.go:425 +0x5c fp=0x14000078f90 sp=0x14000078f60 pc=0x1024b51ac runtime.bgscavenge(0x140000a6000) /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgcscavenge.go:658 +0xac fp=0x14000078fb0 sp=0x14000078f90 pc=0x1024b574c runtime.gcenable.gowrap2() /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:205 +0x28 fp=0x14000078fd0 sp=0x14000078fb0 pc=0x1024ab4b8 runtime.goexit({}) /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/asm_arm64.s:1223 +0x4 fp=0x14000078fd0 sp=0x14000078fd0 pc=0x102508104 created by runtime.gcenable in goroutine 1 /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:205 +0xac goroutine 5 gp=0x14000003c00 m=nil [finalizer wait]: runtime.gopark(0x180007a5c8?, 0x10470fef0?, 0x8?, 0x41?, 0x1c0?) /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/proc.go:435 +0xc8 fp=0x1400007a590 sp=0x1400007a570 pc=0x1024ffd18 runtime.runfinq() /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mfinal.go:196 +0x108 fp=0x1400007a7d0 sp=0x1400007a590 pc=0x1024aa518 runtime.goexit({}) /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/asm_arm64.s:1223 +0x4 fp=0x1400007a7d0 sp=0x1400007a7d0 pc=0x102508104 created by runtime.createfing in goroutine 1 /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mfinal.go:166 +0x80 goroutine 6 gp=0x140001f4700 m=nil [chan receive]: runtime.gopark(0x140002514a0?, 0x14000a21f20?, 0x48?, 0xc7?, 0x1025c88f8?) /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/proc.go:435 +0xc8 fp=0x1400007c6f0 sp=0x1400007c6d0 pc=0x1024ffd18 runtime.chanrecv(0x140000b4310, 0x0, 0x1) /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/chan.go:664 +0x42c fp=0x1400007c770 sp=0x1400007c6f0 pc=0x10249c88c runtime.chanrecv1(0x0?, 0x0?) /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/chan.go:506 +0x14 fp=0x1400007c7a0 sp=0x1400007c770 pc=0x10249c424 runtime.unique_runtime_registerUniqueMapCleanup.func2(...) /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:1796 runtime.unique_runtime_registerUniqueMapCleanup.gowrap1() /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:1799 +0x3c fp=0x1400007c7d0 sp=0x1400007c7a0 pc=0x1024ae73c runtime.goexit({}) /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/asm_arm64.s:1223 +0x4 fp=0x1400007c7d0 sp=0x1400007c7d0 pc=0x102508104 created by unique.runtime_registerUniqueMapCleanup in goroutine 1 /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:1794 +0x78 goroutine 7 gp=0x140001f4e00 m=nil [GC worker (idle)]: runtime.gopark(0x0?, 0x0?, 0x0?, 0x0?, 0x0?) /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/proc.go:435 +0xc8 fp=0x1400007cf10 sp=0x1400007cef0 pc=0x1024ffd18 runtime.gcBgMarkWorker(0x140000b5730) /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:1423 +0xdc fp=0x1400007cfb0 sp=0x1400007cf10 pc=0x1024ad9ac runtime.gcBgMarkStartWorkers.gowrap1() /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:1339 +0x28 fp=0x1400007cfd0 sp=0x1400007cfb0 pc=0x1024ad898 runtime.goexit({}) /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/asm_arm64.s:1223 +0x4 fp=0x1400007cfd0 sp=0x1400007cfd0 pc=0x102508104 created by runtime.gcBgMarkStartWorkers in goroutine 1 /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:1339 +0x140 goroutine 8 gp=0x140001f4fc0 m=nil [GC worker (idle)]: runtime.gopark(0x6bb94e1a4cda?, 0x3?, 0x35?, 0x5?, 0x0?) /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/proc.go:435 +0xc8 fp=0x1400007d710 sp=0x1400007d6f0 pc=0x1024ffd18 runtime.gcBgMarkWorker(0x140000b5730) /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:1423 +0xdc fp=0x1400007d7b0 sp=0x1400007d710 pc=0x1024ad9ac runtime.gcBgMarkStartWorkers.gowrap1() /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:1339 +0x28 fp=0x1400007d7d0 sp=0x1400007d7b0 pc=0x1024ad898 runtime.goexit({}) /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/asm_arm64.s:1223 +0x4 fp=0x1400007d7d0 sp=0x1400007d7d0 pc=0x102508104 created by runtime.gcBgMarkStartWorkers in goroutine 1 /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:1339 +0x140 goroutine 9 gp=0x140001f5180 m=nil [GC worker (idle)]: runtime.gopark(0x10450e0c0?, 0x1?, 0x64?, 0x21?, 0x0?) /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/proc.go:435 +0xc8 fp=0x1400007df10 sp=0x1400007def0 pc=0x1024ffd18 runtime.gcBgMarkWorker(0x140000b5730) /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:1423 +0xdc fp=0x1400007dfb0 sp=0x1400007df10 pc=0x1024ad9ac runtime.gcBgMarkStartWorkers.gowrap1() /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:1339 +0x28 fp=0x1400007dfd0 sp=0x1400007dfb0 pc=0x1024ad898 runtime.goexit({}) /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/asm_arm64.s:1223 +0x4 fp=0x1400007dfd0 sp=0x1400007dfd0 pc=0x102508104 created by runtime.gcBgMarkStartWorkers in goroutine 1 /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:1339 +0x140 goroutine 18 gp=0x14000102380 m=nil [GC worker (idle)]: runtime.gopark(0x6bb94e133d65?, 0x3?, 0xea?, 0x64?, 0x0?) /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/proc.go:435 +0xc8 fp=0x14000076710 sp=0x140000766f0 pc=0x1024ffd18 runtime.gcBgMarkWorker(0x140000b5730) /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:1423 +0xdc fp=0x140000767b0 sp=0x14000076710 pc=0x1024ad9ac runtime.gcBgMarkStartWorkers.gowrap1() /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:1339 +0x28 fp=0x140000767d0 sp=0x140000767b0 pc=0x1024ad898 runtime.goexit({}) /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/asm_arm64.s:1223 +0x4 fp=0x140000767d0 sp=0x140000767d0 pc=0x102508104 created by runtime.gcBgMarkStartWorkers in goroutine 1 /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:1339 +0x140 goroutine 34 gp=0x14000504000 m=nil [GC worker (idle)]: runtime.gopark(0x6bb94e13efc8?, 0x3?, 0x68?, 0x49?, 0x0?) /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/proc.go:435 +0xc8 fp=0x1400050a710 sp=0x1400050a6f0 pc=0x1024ffd18 runtime.gcBgMarkWorker(0x140000b5730) /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:1423 +0xdc fp=0x1400050a7b0 sp=0x1400050a710 pc=0x1024ad9ac runtime.gcBgMarkStartWorkers.gowrap1() /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:1339 +0x28 fp=0x1400050a7d0 sp=0x1400050a7b0 pc=0x1024ad898 runtime.goexit({}) /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/asm_arm64.s:1223 +0x4 fp=0x1400050a7d0 sp=0x1400050a7d0 pc=0x102508104 created by runtime.gcBgMarkStartWorkers in goroutine 1 /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:1339 +0x140 goroutine 10 gp=0x140001f5340 m=nil [GC worker (idle)]: runtime.gopark(0x6bb94e11eab1?, 0x14000112020?, 0x1b?, 0xa?, 0x0?) /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/proc.go:435 +0xc8 fp=0x14000506710 sp=0x140005066f0 pc=0x1024ffd18 runtime.gcBgMarkWorker(0x140000b5730) /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:1423 +0xdc fp=0x140005067b0 sp=0x14000506710 pc=0x1024ad9ac runtime.gcBgMarkStartWorkers.gowrap1() /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:1339 +0x28 fp=0x140005067d0 sp=0x140005067b0 pc=0x1024ad898 runtime.goexit({}) /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/asm_arm64.s:1223 +0x4 fp=0x140005067d0 sp=0x140005067d0 pc=0x102508104 created by runtime.gcBgMarkStartWorkers in goroutine 1 /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:1339 +0x140 goroutine 19 gp=0x14000102540 m=nil [GC worker (idle)]: runtime.gopark(0x6bb50fbb4f9a?, 0x3?, 0xf4?, 0x58?, 0x0?) /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/proc.go:435 +0xc8 fp=0x14000076f10 sp=0x14000076ef0 pc=0x1024ffd18 runtime.gcBgMarkWorker(0x140000b5730) /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:1423 +0xdc fp=0x14000076fb0 sp=0x14000076f10 pc=0x1024ad9ac runtime.gcBgMarkStartWorkers.gowrap1() /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:1339 +0x28 fp=0x14000076fd0 sp=0x14000076fb0 pc=0x1024ad898 runtime.goexit({}) /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/asm_arm64.s:1223 +0x4 fp=0x14000076fd0 sp=0x14000076fd0 pc=0x102508104 created by runtime.gcBgMarkStartWorkers in goroutine 1 /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:1339 +0x140 goroutine 11 gp=0x140001f5500 m=nil [GC worker (idle)]: runtime.gopark(0x6bb94e16fe55?, 0x1?, 0x75?, 0xc5?, 0x0?) /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/proc.go:435 +0xc8 fp=0x14000506f10 sp=0x14000506ef0 pc=0x1024ffd18 runtime.gcBgMarkWorker(0x140000b5730) /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:1423 +0xdc fp=0x14000506fb0 sp=0x14000506f10 pc=0x1024ad9ac runtime.gcBgMarkStartWorkers.gowrap1() /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:1339 +0x28 fp=0x14000506fd0 sp=0x14000506fb0 pc=0x1024ad898 runtime.goexit({}) /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/asm_arm64.s:1223 +0x4 fp=0x14000506fd0 sp=0x14000506fd0 pc=0x102508104 created by runtime.gcBgMarkStartWorkers in goroutine 1 /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:1339 +0x140 goroutine 12 gp=0x140001f56c0 m=nil [GC worker (idle)]: runtime.gopark(0x10450e0c0?, 0x1?, 0xb5?, 0x4a?, 0x0?) /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/proc.go:435 +0xc8 fp=0x14000507710 sp=0x140005076f0 pc=0x1024ffd18 runtime.gcBgMarkWorker(0x140000b5730) /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:1423 +0xdc fp=0x140005077b0 sp=0x14000507710 pc=0x1024ad9ac runtime.gcBgMarkStartWorkers.gowrap1() /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:1339 +0x28 fp=0x140005077d0 sp=0x140005077b0 pc=0x1024ad898 runtime.goexit({}) /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/asm_arm64.s:1223 +0x4 fp=0x140005077d0 sp=0x140005077d0 pc=0x102508104 created by runtime.gcBgMarkStartWorkers in goroutine 1 /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:1339 +0x140 goroutine 13 gp=0x140001f5880 m=nil [GC worker (idle)]: runtime.gopark(0x6bb94e11cd12?, 0x1?, 0x57?, 0x53?, 0x0?) /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/proc.go:435 +0xc8 fp=0x14000507f10 sp=0x14000507ef0 pc=0x1024ffd18 runtime.gcBgMarkWorker(0x140000b5730) /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:1423 +0xdc fp=0x14000507fb0 sp=0x14000507f10 pc=0x1024ad9ac runtime.gcBgMarkStartWorkers.gowrap1() /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:1339 +0x28 fp=0x14000507fd0 sp=0x14000507fb0 pc=0x1024ad898 runtime.goexit({}) /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/asm_arm64.s:1223 +0x4 fp=0x14000507fd0 sp=0x14000507fd0 pc=0x102508104 created by runtime.gcBgMarkStartWorkers in goroutine 1 /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:1339 +0x140 goroutine 35 gp=0x140005041c0 m=nil [GC worker (idle)]: runtime.gopark(0x6bb94e16fb91?, 0x3?, 0xa0?, 0x78?, 0x0?) /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/proc.go:435 +0xc8 fp=0x1400050af10 sp=0x1400050aef0 pc=0x1024ffd18 runtime.gcBgMarkWorker(0x140000b5730) /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:1423 +0xdc fp=0x1400050afb0 sp=0x1400050af10 pc=0x1024ad9ac runtime.gcBgMarkStartWorkers.gowrap1() /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:1339 +0x28 fp=0x1400050afd0 sp=0x1400050afb0 pc=0x1024ad898 runtime.goexit({}) /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/asm_arm64.s:1223 +0x4 fp=0x1400050afd0 sp=0x1400050afd0 pc=0x102508104 created by runtime.gcBgMarkStartWorkers in goroutine 1 /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:1339 +0x140 goroutine 14 gp=0x140001f5a40 m=nil [GC worker (idle)]: runtime.gopark(0x6bb94e1ab3e1?, 0x1?, 0xfe?, 0xa6?, 0x0?) /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/proc.go:435 +0xc8 fp=0x14000508710 sp=0x140005086f0 pc=0x1024ffd18 runtime.gcBgMarkWorker(0x140000b5730) /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:1423 +0xdc fp=0x140005087b0 sp=0x14000508710 pc=0x1024ad9ac runtime.gcBgMarkStartWorkers.gowrap1() /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:1339 +0x28 fp=0x140005087d0 sp=0x140005087b0 pc=0x1024ad898 runtime.goexit({}) /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/asm_arm64.s:1223 +0x4 fp=0x140005087d0 sp=0x140005087d0 pc=0x102508104 created by runtime.gcBgMarkStartWorkers in goroutine 1 /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:1339 +0x140 goroutine 36 gp=0x14000504380 m=nil [GC worker (idle)]: runtime.gopark(0x6bb94e1a7684?, 0x1?, 0x7c?, 0x8?, 0x0?) /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/proc.go:435 +0xc8 fp=0x1400050b710 sp=0x1400050b6f0 pc=0x1024ffd18 runtime.gcBgMarkWorker(0x140000b5730) /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:1423 +0xdc fp=0x1400050b7b0 sp=0x1400050b710 pc=0x1024ad9ac runtime.gcBgMarkStartWorkers.gowrap1() /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:1339 +0x28 fp=0x1400050b7d0 sp=0x1400050b7b0 pc=0x1024ad898 runtime.goexit({}) /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/asm_arm64.s:1223 +0x4 fp=0x1400050b7d0 sp=0x1400050b7d0 pc=0x102508104 created by runtime.gcBgMarkStartWorkers in goroutine 1 /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:1339 +0x140 goroutine 37 gp=0x14000504540 m=nil [GC worker (idle)]: runtime.gopark(0x10450e0c0?, 0x3?, 0x84?, 0x68?, 0x0?) /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/proc.go:435 +0xc8 fp=0x1400050bf10 sp=0x1400050bef0 pc=0x1024ffd18 runtime.gcBgMarkWorker(0x140000b5730) /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:1423 +0xdc fp=0x1400050bfb0 sp=0x1400050bf10 pc=0x1024ad9ac runtime.gcBgMarkStartWorkers.gowrap1() /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:1339 +0x28 fp=0x1400050bfd0 sp=0x1400050bfb0 pc=0x1024ad898 runtime.goexit({}) /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/asm_arm64.s:1223 +0x4 fp=0x1400050bfd0 sp=0x1400050bfd0 pc=0x102508104 created by runtime.gcBgMarkStartWorkers in goroutine 1 /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/mgc.go:1339 +0x140 goroutine 16 gp=0x14000682e00 m=nil [select]: runtime.gopark(0x14000a83a50?, 0x2?, 0x28?, 0x6?, 0x14000a8381c?) /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/proc.go:435 +0xc8 fp=0x14000a83640 sp=0x14000a83620 pc=0x1024ffd18 runtime.selectgo(0x14000a83a50, 0x14000a83818, 0x14000626300?, 0x0, 0x1?, 0x1) /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/select.go:351 +0x6c4 fp=0x14000a83770 sp=0x14000a83640 pc=0x1024df954 github.com/ollama/ollama/runner/ollamarunner.(*Server).completion(0x1400023b0e0, {0x103a7ac60, 0x1400016c0e0}, 0x140004c2140) /Users/runner/work/ollama/ollama/runner/ollamarunner/runner.go:956 +0xa5c fp=0x14000a83aa0 sp=0x14000a83770 pc=0x102a54a2c github.com/ollama/ollama/runner/ollamarunner.(*Server).completion-fm({0x103a7ac60?, 0x1400016c0e0?}, 0x14000a87b28?) <autogenerated>:1 +0x40 fp=0x14000a83ad0 sp=0x14000a83aa0 pc=0x102a591e0 net/http.HandlerFunc.ServeHTTP(0x140005258c0?, {0x103a7ac60?, 0x1400016c0e0?}, 0x14000a87b10?) /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/net/http/server.go:2294 +0x38 fp=0x14000a83b00 sp=0x14000a83ad0 pc=0x1027b9a78 net/http.(*ServeMux).ServeHTTP(0x10?, {0x103a7ac60, 0x1400016c0e0}, 0x140004c2140) /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/net/http/server.go:2822 +0x1b4 fp=0x14000a83b50 sp=0x14000a83b00 pc=0x1027bb604 net/http.serverHandler.ServeHTTP({0x103a76f90?}, {0x103a7ac60?, 0x1400016c0e0?}, 0x1?) /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/net/http/server.go:3301 +0xbc fp=0x14000a83b80 sp=0x14000a83b50 pc=0x1027d72ec net/http.(*conn).serve(0x140004d8480, {0x103a7d328, 0x14000234240}) /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/net/http/server.go:2102 +0x52c fp=0x14000a83fa0 sp=0x14000a83b80 pc=0x1027b821c net/http.(*Server).Serve.gowrap3() /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/net/http/server.go:3454 +0x30 fp=0x14000a83fd0 sp=0x14000a83fa0 pc=0x1027bd3e0 runtime.goexit({}) /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/asm_arm64.s:1223 +0x4 fp=0x14000a83fd0 sp=0x14000a83fd0 pc=0x102508104 created by net/http.(*Server).Serve in goroutine 1 /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/net/http/server.go:3454 +0x3d8 goroutine 543 gp=0x14000103dc0 m=nil [IO wait]: runtime.gopark(0xffffffffffffffff?, 0xffffffffffffffff?, 0x23?, 0x0?, 0x102523920?) /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/proc.go:435 +0xc8 fp=0x14000187d80 sp=0x14000187d60 pc=0x1024ffd18 runtime.netpollblock(0x0?, 0x0?, 0x0?) /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/netpoll.go:575 +0x158 fp=0x14000187dc0 sp=0x14000187d80 pc=0x1024c5778 internal/poll.runtime_pollWait(0x12d8333f8, 0x72) /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/netpoll.go:351 +0xa0 fp=0x14000187df0 sp=0x14000187dc0 pc=0x1024feed0 internal/poll.(*pollDesc).wait(0x14000627600?, 0x14000234701?, 0x0) /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/internal/poll/fd_poll_runtime.go:84 +0x28 fp=0x14000187e20 sp=0x14000187df0 pc=0x10257fc88 internal/poll.(*pollDesc).waitRead(...) /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/internal/poll/fd_poll_runtime.go:89 internal/poll.(*FD).Read(0x14000627600, {0x14000234701, 0x1, 0x1}) /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/internal/poll/fd_unix.go:165 +0x1fc fp=0x14000187ec0 sp=0x14000187e20 pc=0x102580f3c net.(*netFD).Read(0x14000627600, {0x14000234701?, 0x14000187f58?, 0x1027b2c94?}) /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/net/fd_posix.go:55 +0x28 fp=0x14000187f10 sp=0x14000187ec0 pc=0x1025f2cf8 net.(*conn).Read(0x14000120628, {0x14000234701?, 0xff000000ff000000?, 0xff000000ff000000?}) /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/net/net.go:194 +0x34 fp=0x14000187f60 sp=0x14000187f10 pc=0x1025ffbc4 net/http.(*connReader).backgroundRead(0x140002346f0) /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/net/http/server.go:690 +0x40 fp=0x14000187fb0 sp=0x14000187f60 pc=0x1027b2b90 net/http.(*connReader).startBackgroundRead.gowrap2() /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/net/http/server.go:686 +0x28 fp=0x14000187fd0 sp=0x14000187fb0 pc=0x1027b2a78 runtime.goexit({}) /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/asm_arm64.s:1223 +0x4 fp=0x14000187fd0 sp=0x14000187fd0 pc=0x102508104 created by net/http.(*connReader).startBackgroundRead in goroutine 16 /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/net/http/server.go:686 +0xc4 goroutine 1529 gp=0x14000602e00 m=4 mp=0x14000081808 [syscall]: runtime.cgocall(0x1031e8c1c, 0x140011a3a78) /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/cgocall.go:167 +0x44 fp=0x140011a3a40 sp=0x140011a3a00 pc=0x1024fc7f4 github.com/ollama/ollama/ml/backend/ggml._Cfunc_ggml_backend_sched_synchronize(0x13200aa00) _cgo_gotypes.go:1044 +0x30 fp=0x140011a3a70 sp=0x140011a3a40 pc=0x1029606b0 github.com/ollama/ollama/ml/backend/ggml.(*Context).ComputeWithNotify.func4.1(...) /Users/runner/work/ollama/ollama/ml/backend/ggml/ggml.go:833 github.com/ollama/ollama/ml/backend/ggml.(*Context).ComputeWithNotify.func4() /Users/runner/work/ollama/ollama/ml/backend/ggml/ggml.go:833 +0x64 fp=0x140011a3ac0 sp=0x140011a3a70 pc=0x10296ab04 github.com/ollama/ollama/ml/backend/ggml.(*Tensor).Floats(0x1400038a8b8) /Users/runner/work/ollama/ollama/ml/backend/ggml/ggml.go:1065 +0x94 fp=0x140011a3b50 sp=0x140011a3ac0 pc=0x10296c564 github.com/ollama/ollama/runner/ollamarunner.(*Server).computeBatch(0x1400023b0e0, {0x1b8, {0x103a89550, 0x1400052c080}, {0x103a966c0, 0x1400038a8b8}, {0x1400007e070, 0x1, 0x1}, {{0x103a966c0, ...}, ...}, ...}) /Users/runner/work/ollama/ollama/runner/ollamarunner/runner.go:723 +0x714 fp=0x140011a3ed0 sp=0x140011a3b50 pc=0x102a52d94 github.com/ollama/ollama/runner/ollamarunner.(*Server).run.gowrap1() /Users/runner/work/ollama/ollama/runner/ollamarunner/runner.go:459 +0x5c fp=0x140011a3fd0 sp=0x140011a3ed0 pc=0x102a5117c runtime.goexit({}) /Users/runner/hostedtoolcache/go/1.24.1/arm64/src/runtime/asm_arm64.s:1223 +0x4 fp=0x140011a3fd0 sp=0x140011a3fd0 pc=0x102508104 created by github.com/ollama/ollama/runner/ollamarunner.(*Server).run in goroutine 15 /Users/runner/work/ollama/ollama/runner/ollamarunner/runner.go:459 +0x22c r0 0x0 r1 0x0 r2 0x0 r3 0x0 r4 0x192507910 r5 0x1712cdc40 r6 0x38 r7 0x0 r8 0x42e83c7e5a9e6cd8 r9 0x42e83c7f2bb29cd8 r10 0x3bb r11 0x6 r12 0x6 r13 0x1712cd972 r14 0x102097300 r15 0x1 r16 0x148 r17 0x2045aa2c0 r18 0x0 r19 0x6 r20 0x3203 r21 0x1712cf0e0 r22 0x140092defb0 r23 0x8 r24 0x20000 r25 0x80 r26 0x1400118dde0 r27 0x828 r28 0x140005048c0 r29 0x1712ce530 lr 0x1925f7f70 sp 0x1712ce510 pc 0x1925bf720 fault 0x1925bf720 time=2026-02-19T17:00:46.955+01:00 level=ERROR source=server.go:304 msg="llama runner terminated" error="exit status 2" [GIN] 2026/02/19 - 17:00:46 | 500 | 19.285960042s | 10.210.21.37 | POST "/api/generate" here is the startup stuff from my log: time=2026-02-18T08:10:12.775+01:00 level=INFO source=routes.go:1663 msg="server config" env="map[HTTPS_PROXY: HTTP_PROXY: NO_PROXY: OLLAMA_CONTEXT_LENGTH:16384 OLLAMA_DEBUG:INFO OLLAMA_EDITOR: OLLAMA_FLASH_ATTENTION:false OLLAMA_GPU_OVERHEAD:0 OLLAMA_HOST:http://0.0.0.0:11434 OLLAMA_KEEP_ALIVE:5m0s OLLAMA_KV_CACHE_TYPE: OLLAMA_LLM_LIBRARY: OLLAMA_LOAD_TIMEOUT:5m0s OLLAMA_MAX_LOADED_MODELS:0 OLLAMA_MAX_QUEUE:512 OLLAMA_MODELS:/Users/macosx/.ollama/models OLLAMA_MULTIUSER_CACHE:false OLLAMA_NEW_ENGINE:false OLLAMA_NOHISTORY:false OLLAMA_NOPRUNE:false OLLAMA_NO_CLOUD:true OLLAMA_NUM_PARALLEL:1 OLLAMA_ORIGINS:[http://localhost https://localhost http://localhost:* https://localhost:* http://127.0.0.1 https://127.0.0.1 http://127.0.0.1:* https://127.0.0.1:* http://0.0.0.0 https://0.0.0.0 http://0.0.0.0:* https://0.0.0.0:* app://* file://* tauri://* vscode-webview://* vscode-file://*] OLLAMA_REMOTES:[ollama.com] OLLAMA_SCHED_SPREAD:false http_proxy: https_proxy: no_proxy:]" ...
GiteaMirror commented 2026-04-29 09:47:16 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Feb 19, 2026):

Does it fail with context 8192?

<!-- gh-comment-id:3928314540 --> @rick-github commented on GitHub (Feb 19, 2026): Does it fail with context 8192?
GiteaMirror commented 2026-04-29 09:47:18 -05:00
Author
Owner
Copy Link

@daboe01 commented on GitHub (Feb 20, 2026):

no, works indeed :-)

<!-- gh-comment-id:3932148972 --> @daboe01 commented on GitHub (Feb 20, 2026): no, works indeed :-)
GiteaMirror commented 2026-04-29 09:47:19 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Feb 20, 2026):

I suspect this is the same as #14171, closing as a dupe. I think the root cause of this will be fixed in the next vendor sync.

<!-- gh-comment-id:3935336172 --> @rick-github commented on GitHub (Feb 20, 2026): I suspect this is the same as #14171, closing as a dupe. I think the root cause of this will be fixed in the next vendor sync.
GiteaMirror commented 2026-04-29 09:47:19 -05:00
Author
Owner
Copy Link

@daboe01 commented on GitHub (Feb 20, 2026):

thank you!

<!-- gh-comment-id:3936295975 --> @daboe01 commented on GitHub (Feb 20, 2026): thank you!
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#55832
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?