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 #12238] Apple M2 Max causes panic: error computing ggml graph: -1 with gemma3:latest and qwen2.5vl:latest #54655

New Issue
Closed
opened 2026-04-29 06:46:29 -05:00 by GiteaMirror · 1 comment
GiteaMirror commented 2026-04-29 06:46:29 -05:00
Owner
Copy Link

Originally created by @moncapitaine on GitHub (Sep 10, 2025).
Original GitHub issue: https://github.com/ollama/ollama/issues/12238

What is the issue?

This happens around 60% of the tries. It does not matter if I try with /api/chat or /api/generate or if I stream or not.

I totally fresh-installed ollama 0.11.10 and removed the old directories as described here: https://github.com/ollama/ollama/blob/main/docs/macos.md

My main concern is about not being able to reproduce it. I tried different ENV settings like GPU=0, restricting MAX RAM etc. . Nothing seems to help.

I had this problem with several images, smaller and bigger ones. I only mention images that work from time to time.

Typical last "done" chunk is:

1. **Brand Name**: The brand name is "...
{
  "message": {
    "role": "assistant",
    "content": ""
  },
  "done": true,
  "total_duration": 30343321958,
  "load_duration": 73879083,
  "prompt_eval_count": 1307,
  "prompt_eval_duration": 22409707625,
  "eval_count": 391,
  "eval_duration": 7857276084
}

This problem seems not to happen if I query the same model with the same image using the OLLAMA Chat UI

Both environments (app and cli serve) seem to be identical:

HTTPS_PROXY: 
HTTP_PROXY: 
NO_PROXY: 
OLLAMA_CONTEXT_LENGTH:4096 
OLLAMA_DEBUG:INFO 
OLLAMA_FLASH_ATTENTION:false 
OLLAMA_GPU_OVERHEAD:0 
OLLAMA_HOST:http://127.0.0.1:11434 
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/michael/.ollama/models 
OLLAMA_MULTIUSER_CACHE:false 
OLLAMA_NEW_ENGINE:false 
OLLAMA_NEW_ESTIMATES:false 
OLLAMA_NOHISTORY:false 
OLLAMA_NOPRUNE:false 
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_SCHED_SPREAD:false 
http_proxy: 
https_proxy: 
no_proxy:

Relevant log output

Test with qwen2.5vl:latest

time=2025-09-10T09:44:52.607+02:00 level=INFO source=server.go:398 msg="starting runner" cmd="/Applications/Ollama.app/Contents/Resources/ollama runner --ollama-engine --model /Users/michael/.ollama/models/blobs/sha256-a99b7f834d754b88f122d865f32758ba9f0994a83f8363df2c1e71c17605a025 --port 54762"
time=2025-09-10T09:44:52.612+02:00 level=INFO source=server.go:503 msg="system memory" total="96.0 GiB" free="70.1 GiB" free_swap="0 B"
time=2025-09-10T09:44:52.614+02:00 level=INFO source=memory.go:36 msg="new model will fit in available VRAM across minimum required GPUs, loading" model=/Users/michael/.ollama/models/blobs/sha256-a99b7f834d754b88f122d865f32758ba9f0994a83f8363df2c1e71c17605a025 library=metal parallel=1 required="8.0 GiB" gpus=1
time=2025-09-10T09:44:52.615+02:00 level=INFO source=server.go:543 msg=offload library=metal layers.requested=-1 layers.model=29 layers.offload=29 layers.split=[29] memory.available="[72.0 GiB]" memory.gpu_overhead="0 B" memory.required.full="8.0 GiB" memory.required.partial="8.0 GiB" memory.required.kv="224.0 MiB" memory.required.allocations="[8.0 GiB]" memory.weights.total="4.1 GiB" memory.weights.repeating="3.7 GiB" memory.weights.nonrepeating="426.4 MiB" memory.graph.full="261.3 MiB" memory.graph.partial="261.3 MiB" projector.weights="1.2 GiB" projector.graph="1.6 GiB"
time=2025-09-10T09:44:52.620+02:00 level=INFO source=runner.go:1251 msg="starting ollama engine"
time=2025-09-10T09:44:52.621+02:00 level=INFO source=runner.go:1286 msg="Server listening on 127.0.0.1:54762"
time=2025-09-10T09:44:52.626+02:00 level=INFO source=runner.go:1170 msg=load request="{Operation:commit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:false KvSize:4096 KvCacheType: NumThreads:8 GPULayers:29[ID:0 Layers:29(0..28)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2025-09-10T09:44:52.644+02:00 level=INFO source=ggml.go:131 msg="" architecture=qwen25vl file_type=Q4_K_M name="" description="" num_tensors=858 num_key_values=36
time=2025-09-10T09:44:52.645+02:00 level=INFO source=ggml.go:104 msg=system Metal.0.EMBED_LIBRARY=1 Metal.0.BF16=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 M2 Max
ggml_metal_load_library: using embedded metal library
ggml_metal_init: GPU name:   Apple M2 Max
ggml_metal_init: GPU family: MTLGPUFamilyApple8  (1008)
ggml_metal_init: GPU family: MTLGPUFamilyCommon3 (3003)
ggml_metal_init: GPU family: MTLGPUFamilyMetal3  (5001)
ggml_metal_init: simdgroup reduction   = true
ggml_metal_init: simdgroup matrix mul. = true
ggml_metal_init: has residency sets    = false
ggml_metal_init: has bfloat            = true
ggml_metal_init: use bfloat            = true
ggml_metal_init: hasUnifiedMemory      = true
ggml_metal_init: recommendedMaxWorkingSetSize  = 77309.41 MB
time=2025-09-10T09:44:52.863+02:00 level=INFO source=ggml.go:487 msg="offloading 28 repeating layers to GPU"
time=2025-09-10T09:44:52.863+02:00 level=INFO source=ggml.go:493 msg="offloading output layer to GPU"
time=2025-09-10T09:44:52.863+02:00 level=INFO source=ggml.go:498 msg="offloaded 29/29 layers to GPU"
time=2025-09-10T09:44:52.863+02:00 level=INFO source=backend.go:310 msg="model weights" device=Metal size="5.3 GiB"
time=2025-09-10T09:44:52.863+02:00 level=INFO source=backend.go:315 msg="model weights" device=CPU size="292.4 MiB"
time=2025-09-10T09:44:52.863+02:00 level=INFO source=backend.go:321 msg="kv cache" device=Metal size="224.0 MiB"
time=2025-09-10T09:44:52.863+02:00 level=INFO source=backend.go:332 msg="compute graph" device=Metal size="1.7 GiB"
time=2025-09-10T09:44:52.863+02:00 level=INFO source=backend.go:337 msg="compute graph" device=CPU size="16.8 MiB"
time=2025-09-10T09:44:52.863+02:00 level=INFO source=backend.go:342 msg="total memory" size="7.4 GiB"
time=2025-09-10T09:44:52.863+02:00 level=INFO source=sched.go:473 msg="loaded runners" count=1
time=2025-09-10T09:44:52.863+02:00 level=INFO source=server.go:1250 msg="waiting for llama runner to start responding"
time=2025-09-10T09:44:52.885+02:00 level=INFO source=server.go:1284 msg="waiting for server to become available" status="llm server loading model"
time=2025-09-10T09:44:54.391+02:00 level=INFO source=server.go:1288 msg="llama runner started in 1.78 seconds"
ggml_metal_graph_compute: command buffer 0 failed with status 5
error: Internal Error (0000000e:Internal Error)
panic: error computing ggml graph: -1

goroutine 9 [running]:
github.com/ollama/ollama/ml/backend/ggml.(*Context).ComputeWithNotify(0x140000da040, 0x0?, {0x14000284000, 0x1, 0x152df1601?})
	/Users/runner/work/ollama/ollama/ml/backend/ggml/ggml.go:772 +0x32c
github.com/ollama/ollama/ml/backend/ggml.(*Context).Compute(0x140000da040?, {0x14000284000?, 0x1?, 0x100d58794?})
	/Users/runner/work/ollama/ollama/ml/backend/ggml/ggml.go:762 +0x30
github.com/ollama/ollama/runner/ollamarunner.multimodalStore.getTensor(0x14000b03868?, {0x1020214d0, 0x140001396b0}, {0x102025be0, 0x140000da000}, {0x1020306e8, 0x14000cd09f0}, 0x0)
	/Users/runner/work/ollama/ollama/runner/ollamarunner/multimodal.go:90 +0x234
github.com/ollama/ollama/runner/ollamarunner.multimodalStore.getMultimodal(0x14001786f00, {0x1020214d0, 0x140001396b0}, {0x102025be0, 0x140000da000}, {0x14000ce6d20, 0x1, 0x20?}, 0x0)
	/Users/runner/work/ollama/ollama/runner/ollamarunner/multimodal.go:56 +0xa8
github.com/ollama/ollama/runner/ollamarunner.(*Server).forwardBatch(_, {0x0, {0x102025be0, 0x14000ce20c0}, {0x1020306e8, 0x14000c62948}, {0x1400028b000, 0x16, 0x20}, {{0x1020306e8, ...}, ...}, ...})
	/Users/runner/work/ollama/ollama/runner/ollamarunner/runner.go:543 +0xbbc
github.com/ollama/ollama/runner/ollamarunner.(*Server).run(0x14000208f00, {0x10201c6b0, 0x140000d3ea0})
	/Users/runner/work/ollama/ollama/runner/ollamarunner/runner.go:420 +0x15c
created by github.com/ollama/ollama/runner/ollamarunner.Execute in goroutine 1
	/Users/runner/work/ollama/ollama/runner/ollamarunner/runner.go:1265 +0x470

Test with gemma3:latest

[GIN] 2025/09/10 - 18:55:53 | 200 | 16.873529708s |       127.0.0.1 | POST     "/api/chat"
time=2025-09-10T18:56:27.597+02:00 level=INFO source=server.go:398 msg="starting runner" cmd="/Applications/Ollama.app/Contents/Resources/ollama runner --ollama-engine --model /Users/michael/.ollama/models/blobs/sha256-aeda25e63ebd698fab8638ffb778e68bed908b960d39d0becc650fa981609d25 --port 53667"
time=2025-09-10T18:56:27.602+02:00 level=INFO source=server.go:503 msg="system memory" total="96.0 GiB" free="67.1 GiB" free_swap="0 B"
time=2025-09-10T18:56:27.603+02:00 level=INFO source=memory.go:36 msg="new model will fit in available VRAM across minimum required GPUs, loading" model=/Users/michael/.ollama/models/blobs/sha256-aeda25e63ebd698fab8638ffb778e68bed908b960d39d0becc650fa981609d25 library=metal parallel=1 required="5.4 GiB" gpus=1
time=2025-09-10T18:56:27.604+02:00 level=INFO source=server.go:543 msg=offload library=metal layers.requested=-1 layers.model=35 layers.offload=35 layers.split=[35] memory.available="[72.0 GiB]" memory.gpu_overhead="0 B" memory.required.full="5.4 GiB" memory.required.partial="5.4 GiB" memory.required.kv="254.0 MiB" memory.required.allocations="[5.4 GiB]" memory.weights.total="2.3 GiB" memory.weights.repeating="1.8 GiB" memory.weights.nonrepeating="525.0 MiB" memory.graph.full="517.0 MiB" memory.graph.partial="517.0 MiB" projector.weights="795.9 MiB" projector.graph="1.0 GiB"
time=2025-09-10T18:56:27.610+02:00 level=INFO source=runner.go:1251 msg="starting ollama engine"
time=2025-09-10T18:56:27.610+02:00 level=INFO source=runner.go:1286 msg="Server listening on 127.0.0.1:53667"
time=2025-09-10T18:56:27.616+02:00 level=INFO source=runner.go:1170 msg=load request="{Operation:commit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:false KvSize:4096 KvCacheType: NumThreads:8 GPULayers:35[ID:0 Layers:35(0..34)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2025-09-10T18:56:27.656+02:00 level=INFO source=ggml.go:131 msg="" architecture=gemma3 file_type=Q4_K_M name="" description="" num_tensors=883 num_key_values=36
time=2025-09-10T18:56:27.657+02:00 level=INFO source=ggml.go:104 msg=system Metal.0.EMBED_LIBRARY=1 Metal.0.BF16=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 M2 Max
ggml_metal_load_library: using embedded metal library
ggml_metal_init: GPU name:   Apple M2 Max
ggml_metal_init: GPU family: MTLGPUFamilyApple8  (1008)
ggml_metal_init: GPU family: MTLGPUFamilyCommon3 (3003)
ggml_metal_init: GPU family: MTLGPUFamilyMetal3  (5001)
ggml_metal_init: simdgroup reduction   = true
ggml_metal_init: simdgroup matrix mul. = true
ggml_metal_init: has residency sets    = false
ggml_metal_init: has bfloat            = true
ggml_metal_init: use bfloat            = true
ggml_metal_init: hasUnifiedMemory      = true
ggml_metal_init: recommendedMaxWorkingSetSize  = 77309.41 MB
time=2025-09-10T18:56:27.843+02:00 level=INFO source=ggml.go:487 msg="offloading 34 repeating layers to GPU"
time=2025-09-10T18:56:27.843+02:00 level=INFO source=ggml.go:493 msg="offloading output layer to GPU"
time=2025-09-10T18:56:27.843+02:00 level=INFO source=ggml.go:498 msg="offloaded 35/35 layers to GPU"
time=2025-09-10T18:56:27.844+02:00 level=INFO source=backend.go:310 msg="model weights" device=Metal size="3.1 GiB"
time=2025-09-10T18:56:27.844+02:00 level=INFO source=backend.go:315 msg="model weights" device=CPU size="525.0 MiB"
time=2025-09-10T18:56:27.844+02:00 level=INFO source=backend.go:321 msg="kv cache" device=Metal size="254.0 MiB"
time=2025-09-10T18:56:27.844+02:00 level=INFO source=backend.go:332 msg="compute graph" device=Metal size="1.1 GiB"
time=2025-09-10T18:56:27.844+02:00 level=INFO source=backend.go:337 msg="compute graph" device=CPU size="5.0 MiB"
time=2025-09-10T18:56:27.844+02:00 level=INFO source=backend.go:342 msg="total memory" size="5.0 GiB"
time=2025-09-10T18:56:27.844+02:00 level=INFO source=sched.go:473 msg="loaded runners" count=1
time=2025-09-10T18:56:27.844+02:00 level=INFO source=server.go:1250 msg="waiting for llama runner to start responding"
time=2025-09-10T18:56:27.848+02:00 level=INFO source=server.go:1284 msg="waiting for server to become available" status="llm server loading model"
time=2025-09-10T18:56:28.099+02:00 level=INFO source=server.go:1288 msg="llama runner started in 0.50 seconds"
[GIN] 2025/09/10 - 18:56:40 | 200 | 12.656067458s |       127.0.0.1 | POST     "/api/chat"
[GIN] 2025/09/10 - 18:56:54 | 200 | 14.497685459s |       127.0.0.1 | POST     "/api/chat"
[GIN] 2025/09/10 - 18:57:07 | 200 | 12.689051333s |       127.0.0.1 | POST     "/api/chat"
ggml_metal_graph_compute: command buffer 0 failed with status 5
error: Internal Error (0000000e:Internal Error)
panic: error computing ggml graph: -1

goroutine 12 [running]:
github.com/ollama/ollama/ml/backend/ggml.(*Context).ComputeWithNotify(0x140017f4440, 0x102904778?, {0x14001832040, 0x1, 0x134473601?})
	/Users/runner/work/ollama/ollama/ml/backend/ggml/ggml.go:772 +0x32c
github.com/ollama/ollama/ml/backend/ggml.(*Context).Compute(0x140017f4440?, {0x14001832040?, 0x1?, 0x14000052008?})
	/Users/runner/work/ollama/ollama/ml/backend/ggml/ggml.go:762 +0x30
github.com/ollama/ollama/runner/ollamarunner.multimodalStore.getTensor(0x14001649868?, {0x10370d4d0, 0x140000d96b0}, {0x103711be0, 0x140017f4400}, {0x10371c6e8, 0x14001847bf0}, 0x0)
	/Users/runner/work/ollama/ollama/runner/ollamarunner/multimodal.go:90 +0x234
github.com/ollama/ollama/runner/ollamarunner.multimodalStore.getMultimodal(0x1400219c210, {0x10370d4d0, 0x140000d96b0}, {0x103711be0, 0x140017f4400}, {0x140002863c0, 0x1, 0x2?}, 0x0)
	/Users/runner/work/ollama/ollama/runner/ollamarunner/multimodal.go:56 +0xa8
github.com/ollama/ollama/runner/ollamarunner.(*Server).forwardBatch(_, {0x360, {0x103711be0, 0x140018b2000}, {0x10371c6e8, 0x140018d4348}, {0x14000076068, 0x1, 0x1}, {{0x10371c6e8, ...}, ...}, ...})
	/Users/runner/work/ollama/ollama/runner/ollamarunner/runner.go:543 +0xbbc
github.com/ollama/ollama/runner/ollamarunner.(*Server).run(0x140001541e0, {0x1037086b0, 0x1400037c190})
	/Users/runner/work/ollama/ollama/runner/ollamarunner/runner.go:420 +0x15c
created by github.com/ollama/ollama/runner/ollamarunner.Execute in goroutine 1
	/Users/runner/work/ollama/ollama/runner/ollamarunner/runner.go:1265 +0x470
time=2025-09-10T18:57:15.333+02:00 level=ERROR source=server.go:1458 msg="post predict" error="Post \"http://127.0.0.1:53667/completion\": EOF"
time=2025-09-10T18:57:15.333+02:00 level=ERROR source=server.go:424 msg="llama runner terminated" error="exit status 2"

OS

macOS

GPU

Apple

CPU

Apple

Ollama version

0.11.10

Originally created by @moncapitaine on GitHub (Sep 10, 2025). Original GitHub issue: https://github.com/ollama/ollama/issues/12238 ### What is the issue? This happens around 60% of the tries. It does not matter if I try with /api/chat or /api/generate or if I stream or not. I totally fresh-installed ollama 0.11.10 and removed the old directories as described here: https://github.com/ollama/ollama/blob/main/docs/macos.md My main concern is about not being able to reproduce it. I tried different ENV settings like GPU=0, restricting MAX RAM etc. . Nothing seems to help. I had this problem with several images, smaller and bigger ones. I only mention images that work from time to time. Typical last "done" chunk is: ```json 1. **Brand Name**: The brand name is "... { "message": { "role": "assistant", "content": "" }, "done": true, "total_duration": 30343321958, "load_duration": 73879083, "prompt_eval_count": 1307, "prompt_eval_duration": 22409707625, "eval_count": 391, "eval_duration": 7857276084 } ``` **This problem seems not to happen if I query the same model with the same image using the OLLAMA Chat UI** Both environments (app and cli serve) seem to be identical: ``` HTTPS_PROXY: HTTP_PROXY: NO_PROXY: OLLAMA_CONTEXT_LENGTH:4096 OLLAMA_DEBUG:INFO OLLAMA_FLASH_ATTENTION:false OLLAMA_GPU_OVERHEAD:0 OLLAMA_HOST:http://127.0.0.1:11434 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/michael/.ollama/models OLLAMA_MULTIUSER_CACHE:false OLLAMA_NEW_ENGINE:false OLLAMA_NEW_ESTIMATES:false OLLAMA_NOHISTORY:false OLLAMA_NOPRUNE:false 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_SCHED_SPREAD:false http_proxy: https_proxy: no_proxy: ``` ### Relevant log output #### Test with qwen2.5vl:latest ```shell time=2025-09-10T09:44:52.607+02:00 level=INFO source=server.go:398 msg="starting runner" cmd="/Applications/Ollama.app/Contents/Resources/ollama runner --ollama-engine --model /Users/michael/.ollama/models/blobs/sha256-a99b7f834d754b88f122d865f32758ba9f0994a83f8363df2c1e71c17605a025 --port 54762" time=2025-09-10T09:44:52.612+02:00 level=INFO source=server.go:503 msg="system memory" total="96.0 GiB" free="70.1 GiB" free_swap="0 B" time=2025-09-10T09:44:52.614+02:00 level=INFO source=memory.go:36 msg="new model will fit in available VRAM across minimum required GPUs, loading" model=/Users/michael/.ollama/models/blobs/sha256-a99b7f834d754b88f122d865f32758ba9f0994a83f8363df2c1e71c17605a025 library=metal parallel=1 required="8.0 GiB" gpus=1 time=2025-09-10T09:44:52.615+02:00 level=INFO source=server.go:543 msg=offload library=metal layers.requested=-1 layers.model=29 layers.offload=29 layers.split=[29] memory.available="[72.0 GiB]" memory.gpu_overhead="0 B" memory.required.full="8.0 GiB" memory.required.partial="8.0 GiB" memory.required.kv="224.0 MiB" memory.required.allocations="[8.0 GiB]" memory.weights.total="4.1 GiB" memory.weights.repeating="3.7 GiB" memory.weights.nonrepeating="426.4 MiB" memory.graph.full="261.3 MiB" memory.graph.partial="261.3 MiB" projector.weights="1.2 GiB" projector.graph="1.6 GiB" time=2025-09-10T09:44:52.620+02:00 level=INFO source=runner.go:1251 msg="starting ollama engine" time=2025-09-10T09:44:52.621+02:00 level=INFO source=runner.go:1286 msg="Server listening on 127.0.0.1:54762" time=2025-09-10T09:44:52.626+02:00 level=INFO source=runner.go:1170 msg=load request="{Operation:commit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:false KvSize:4096 KvCacheType: NumThreads:8 GPULayers:29[ID:0 Layers:29(0..28)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2025-09-10T09:44:52.644+02:00 level=INFO source=ggml.go:131 msg="" architecture=qwen25vl file_type=Q4_K_M name="" description="" num_tensors=858 num_key_values=36 time=2025-09-10T09:44:52.645+02:00 level=INFO source=ggml.go:104 msg=system Metal.0.EMBED_LIBRARY=1 Metal.0.BF16=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 M2 Max ggml_metal_load_library: using embedded metal library ggml_metal_init: GPU name: Apple M2 Max ggml_metal_init: GPU family: MTLGPUFamilyApple8 (1008) ggml_metal_init: GPU family: MTLGPUFamilyCommon3 (3003) ggml_metal_init: GPU family: MTLGPUFamilyMetal3 (5001) ggml_metal_init: simdgroup reduction = true ggml_metal_init: simdgroup matrix mul. = true ggml_metal_init: has residency sets = false ggml_metal_init: has bfloat = true ggml_metal_init: use bfloat = true ggml_metal_init: hasUnifiedMemory = true ggml_metal_init: recommendedMaxWorkingSetSize = 77309.41 MB time=2025-09-10T09:44:52.863+02:00 level=INFO source=ggml.go:487 msg="offloading 28 repeating layers to GPU" time=2025-09-10T09:44:52.863+02:00 level=INFO source=ggml.go:493 msg="offloading output layer to GPU" time=2025-09-10T09:44:52.863+02:00 level=INFO source=ggml.go:498 msg="offloaded 29/29 layers to GPU" time=2025-09-10T09:44:52.863+02:00 level=INFO source=backend.go:310 msg="model weights" device=Metal size="5.3 GiB" time=2025-09-10T09:44:52.863+02:00 level=INFO source=backend.go:315 msg="model weights" device=CPU size="292.4 MiB" time=2025-09-10T09:44:52.863+02:00 level=INFO source=backend.go:321 msg="kv cache" device=Metal size="224.0 MiB" time=2025-09-10T09:44:52.863+02:00 level=INFO source=backend.go:332 msg="compute graph" device=Metal size="1.7 GiB" time=2025-09-10T09:44:52.863+02:00 level=INFO source=backend.go:337 msg="compute graph" device=CPU size="16.8 MiB" time=2025-09-10T09:44:52.863+02:00 level=INFO source=backend.go:342 msg="total memory" size="7.4 GiB" time=2025-09-10T09:44:52.863+02:00 level=INFO source=sched.go:473 msg="loaded runners" count=1 time=2025-09-10T09:44:52.863+02:00 level=INFO source=server.go:1250 msg="waiting for llama runner to start responding" time=2025-09-10T09:44:52.885+02:00 level=INFO source=server.go:1284 msg="waiting for server to become available" status="llm server loading model" time=2025-09-10T09:44:54.391+02:00 level=INFO source=server.go:1288 msg="llama runner started in 1.78 seconds" ggml_metal_graph_compute: command buffer 0 failed with status 5 error: Internal Error (0000000e:Internal Error) panic: error computing ggml graph: -1 goroutine 9 [running]: github.com/ollama/ollama/ml/backend/ggml.(*Context).ComputeWithNotify(0x140000da040, 0x0?, {0x14000284000, 0x1, 0x152df1601?}) /Users/runner/work/ollama/ollama/ml/backend/ggml/ggml.go:772 +0x32c github.com/ollama/ollama/ml/backend/ggml.(*Context).Compute(0x140000da040?, {0x14000284000?, 0x1?, 0x100d58794?}) /Users/runner/work/ollama/ollama/ml/backend/ggml/ggml.go:762 +0x30 github.com/ollama/ollama/runner/ollamarunner.multimodalStore.getTensor(0x14000b03868?, {0x1020214d0, 0x140001396b0}, {0x102025be0, 0x140000da000}, {0x1020306e8, 0x14000cd09f0}, 0x0) /Users/runner/work/ollama/ollama/runner/ollamarunner/multimodal.go:90 +0x234 github.com/ollama/ollama/runner/ollamarunner.multimodalStore.getMultimodal(0x14001786f00, {0x1020214d0, 0x140001396b0}, {0x102025be0, 0x140000da000}, {0x14000ce6d20, 0x1, 0x20?}, 0x0) /Users/runner/work/ollama/ollama/runner/ollamarunner/multimodal.go:56 +0xa8 github.com/ollama/ollama/runner/ollamarunner.(*Server).forwardBatch(_, {0x0, {0x102025be0, 0x14000ce20c0}, {0x1020306e8, 0x14000c62948}, {0x1400028b000, 0x16, 0x20}, {{0x1020306e8, ...}, ...}, ...}) /Users/runner/work/ollama/ollama/runner/ollamarunner/runner.go:543 +0xbbc github.com/ollama/ollama/runner/ollamarunner.(*Server).run(0x14000208f00, {0x10201c6b0, 0x140000d3ea0}) /Users/runner/work/ollama/ollama/runner/ollamarunner/runner.go:420 +0x15c created by github.com/ollama/ollama/runner/ollamarunner.Execute in goroutine 1 /Users/runner/work/ollama/ollama/runner/ollamarunner/runner.go:1265 +0x470 ``` #### Test with gemma3:latest ``` [GIN] 2025/09/10 - 18:55:53 | 200 | 16.873529708s | 127.0.0.1 | POST "/api/chat" time=2025-09-10T18:56:27.597+02:00 level=INFO source=server.go:398 msg="starting runner" cmd="/Applications/Ollama.app/Contents/Resources/ollama runner --ollama-engine --model /Users/michael/.ollama/models/blobs/sha256-aeda25e63ebd698fab8638ffb778e68bed908b960d39d0becc650fa981609d25 --port 53667" time=2025-09-10T18:56:27.602+02:00 level=INFO source=server.go:503 msg="system memory" total="96.0 GiB" free="67.1 GiB" free_swap="0 B" time=2025-09-10T18:56:27.603+02:00 level=INFO source=memory.go:36 msg="new model will fit in available VRAM across minimum required GPUs, loading" model=/Users/michael/.ollama/models/blobs/sha256-aeda25e63ebd698fab8638ffb778e68bed908b960d39d0becc650fa981609d25 library=metal parallel=1 required="5.4 GiB" gpus=1 time=2025-09-10T18:56:27.604+02:00 level=INFO source=server.go:543 msg=offload library=metal layers.requested=-1 layers.model=35 layers.offload=35 layers.split=[35] memory.available="[72.0 GiB]" memory.gpu_overhead="0 B" memory.required.full="5.4 GiB" memory.required.partial="5.4 GiB" memory.required.kv="254.0 MiB" memory.required.allocations="[5.4 GiB]" memory.weights.total="2.3 GiB" memory.weights.repeating="1.8 GiB" memory.weights.nonrepeating="525.0 MiB" memory.graph.full="517.0 MiB" memory.graph.partial="517.0 MiB" projector.weights="795.9 MiB" projector.graph="1.0 GiB" time=2025-09-10T18:56:27.610+02:00 level=INFO source=runner.go:1251 msg="starting ollama engine" time=2025-09-10T18:56:27.610+02:00 level=INFO source=runner.go:1286 msg="Server listening on 127.0.0.1:53667" time=2025-09-10T18:56:27.616+02:00 level=INFO source=runner.go:1170 msg=load request="{Operation:commit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:false KvSize:4096 KvCacheType: NumThreads:8 GPULayers:35[ID:0 Layers:35(0..34)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2025-09-10T18:56:27.656+02:00 level=INFO source=ggml.go:131 msg="" architecture=gemma3 file_type=Q4_K_M name="" description="" num_tensors=883 num_key_values=36 time=2025-09-10T18:56:27.657+02:00 level=INFO source=ggml.go:104 msg=system Metal.0.EMBED_LIBRARY=1 Metal.0.BF16=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 M2 Max ggml_metal_load_library: using embedded metal library ggml_metal_init: GPU name: Apple M2 Max ggml_metal_init: GPU family: MTLGPUFamilyApple8 (1008) ggml_metal_init: GPU family: MTLGPUFamilyCommon3 (3003) ggml_metal_init: GPU family: MTLGPUFamilyMetal3 (5001) ggml_metal_init: simdgroup reduction = true ggml_metal_init: simdgroup matrix mul. = true ggml_metal_init: has residency sets = false ggml_metal_init: has bfloat = true ggml_metal_init: use bfloat = true ggml_metal_init: hasUnifiedMemory = true ggml_metal_init: recommendedMaxWorkingSetSize = 77309.41 MB time=2025-09-10T18:56:27.843+02:00 level=INFO source=ggml.go:487 msg="offloading 34 repeating layers to GPU" time=2025-09-10T18:56:27.843+02:00 level=INFO source=ggml.go:493 msg="offloading output layer to GPU" time=2025-09-10T18:56:27.843+02:00 level=INFO source=ggml.go:498 msg="offloaded 35/35 layers to GPU" time=2025-09-10T18:56:27.844+02:00 level=INFO source=backend.go:310 msg="model weights" device=Metal size="3.1 GiB" time=2025-09-10T18:56:27.844+02:00 level=INFO source=backend.go:315 msg="model weights" device=CPU size="525.0 MiB" time=2025-09-10T18:56:27.844+02:00 level=INFO source=backend.go:321 msg="kv cache" device=Metal size="254.0 MiB" time=2025-09-10T18:56:27.844+02:00 level=INFO source=backend.go:332 msg="compute graph" device=Metal size="1.1 GiB" time=2025-09-10T18:56:27.844+02:00 level=INFO source=backend.go:337 msg="compute graph" device=CPU size="5.0 MiB" time=2025-09-10T18:56:27.844+02:00 level=INFO source=backend.go:342 msg="total memory" size="5.0 GiB" time=2025-09-10T18:56:27.844+02:00 level=INFO source=sched.go:473 msg="loaded runners" count=1 time=2025-09-10T18:56:27.844+02:00 level=INFO source=server.go:1250 msg="waiting for llama runner to start responding" time=2025-09-10T18:56:27.848+02:00 level=INFO source=server.go:1284 msg="waiting for server to become available" status="llm server loading model" time=2025-09-10T18:56:28.099+02:00 level=INFO source=server.go:1288 msg="llama runner started in 0.50 seconds" [GIN] 2025/09/10 - 18:56:40 | 200 | 12.656067458s | 127.0.0.1 | POST "/api/chat" [GIN] 2025/09/10 - 18:56:54 | 200 | 14.497685459s | 127.0.0.1 | POST "/api/chat" [GIN] 2025/09/10 - 18:57:07 | 200 | 12.689051333s | 127.0.0.1 | POST "/api/chat" ggml_metal_graph_compute: command buffer 0 failed with status 5 error: Internal Error (0000000e:Internal Error) panic: error computing ggml graph: -1 goroutine 12 [running]: github.com/ollama/ollama/ml/backend/ggml.(*Context).ComputeWithNotify(0x140017f4440, 0x102904778?, {0x14001832040, 0x1, 0x134473601?}) /Users/runner/work/ollama/ollama/ml/backend/ggml/ggml.go:772 +0x32c github.com/ollama/ollama/ml/backend/ggml.(*Context).Compute(0x140017f4440?, {0x14001832040?, 0x1?, 0x14000052008?}) /Users/runner/work/ollama/ollama/ml/backend/ggml/ggml.go:762 +0x30 github.com/ollama/ollama/runner/ollamarunner.multimodalStore.getTensor(0x14001649868?, {0x10370d4d0, 0x140000d96b0}, {0x103711be0, 0x140017f4400}, {0x10371c6e8, 0x14001847bf0}, 0x0) /Users/runner/work/ollama/ollama/runner/ollamarunner/multimodal.go:90 +0x234 github.com/ollama/ollama/runner/ollamarunner.multimodalStore.getMultimodal(0x1400219c210, {0x10370d4d0, 0x140000d96b0}, {0x103711be0, 0x140017f4400}, {0x140002863c0, 0x1, 0x2?}, 0x0) /Users/runner/work/ollama/ollama/runner/ollamarunner/multimodal.go:56 +0xa8 github.com/ollama/ollama/runner/ollamarunner.(*Server).forwardBatch(_, {0x360, {0x103711be0, 0x140018b2000}, {0x10371c6e8, 0x140018d4348}, {0x14000076068, 0x1, 0x1}, {{0x10371c6e8, ...}, ...}, ...}) /Users/runner/work/ollama/ollama/runner/ollamarunner/runner.go:543 +0xbbc github.com/ollama/ollama/runner/ollamarunner.(*Server).run(0x140001541e0, {0x1037086b0, 0x1400037c190}) /Users/runner/work/ollama/ollama/runner/ollamarunner/runner.go:420 +0x15c created by github.com/ollama/ollama/runner/ollamarunner.Execute in goroutine 1 /Users/runner/work/ollama/ollama/runner/ollamarunner/runner.go:1265 +0x470 time=2025-09-10T18:57:15.333+02:00 level=ERROR source=server.go:1458 msg="post predict" error="Post \"http://127.0.0.1:53667/completion\": EOF" time=2025-09-10T18:57:15.333+02:00 level=ERROR source=server.go:424 msg="llama runner terminated" error="exit status 2" ``` ### OS macOS ### GPU Apple ### CPU Apple ### Ollama version 0.11.10
GiteaMirror added the bug label 2026-04-29 06:46:29 -05:00
GiteaMirror commented 2026-04-29 06:46:31 -05:00
Author
Owner
Copy Link

@jessegross commented on GitHub (Sep 10, 2025):

Thanks for the logs and investigation. There is an open bug for this with #10986 - please focus any discussion there.

<!-- gh-comment-id:3275924215 --> @jessegross commented on GitHub (Sep 10, 2025): Thanks for the logs and investigation. There is an open bug for this with #10986 - please focus any discussion there.
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#54655
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?