github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-06 16:11:34 -05:00
Code Issues 14.5k Packages Projects Releases 120 Wiki Activity

[GH-ISSUE #14615] Error in ollama using open-notebook #35231

New Issue
Closed
opened 2026-04-22 19:37:01 -05:00 by GiteaMirror · 4 comments
GiteaMirror commented 2026-04-22 19:37:01 -05:00
Owner
Copy Link

Originally created by @ScaryBeats01 on GitHub (Mar 4, 2026).
Original GitHub issue: https://github.com/ollama/ollama/issues/14615

What is the issue?

When I use open-notebook, it allways gives error 500 in just 30 seconds on creating a insight.
But if I use the "ollama run" featur, only with commandr-7b:latest gives this errors. I tried qwen3.5:4b, granite3.1-dense:2b, granite4:tiny-h and lfm2.5-thinking without errors.
Look:
PS C:\Users\user> ollama run command-r7b:latest "Summarize this text in 3 bullet points:

INPUT

Neural networks are computing systems inspired by biological neural networks..."
Error: 500 Internal Server Error: llama runner process has terminated: CUDA error
PS C:\Users\user> ollama run command-r7b:latest "Summarize this text in 3 bullet points:

INPUT

Neural networks are computing systems inspired by biological neural networks..."^C
PS C:\Users\user> ollama rm command-r7b:latest
deleted 'command-r7b:latest'
PS C:\Users\user> ollama pull command-r7b:latest
pulling manifest
pulling b32d935e114c: 100% ▕███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████▏ 5.1 GB
pulling 0d8282caa612: 100% ▕███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████▏ 7.2 KB
pulling 945eaa8b1428: 100% ▕███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████▏ 13 KB
pulling d8455b5dce0b: 100% ▕███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████▏ 110 B
pulling 574fdc7616e8: 100% ▕███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████▏ 491 B
verifying sha256 digest
writing manifest
success
PS C:\Users\user> ollama run command-r7b:latest "Summarize this text in 3 bullet points:

INPUT

Neural networks are computing systems inspired by biological neural networks..."
Error: 500 Internal Server Error: llama runner process has terminated: CUDA error
PS C:\Users\user> ollama ps
NAME ID SIZE PROCESSOR CONTEXT UNTIL

Relevant log output

print_info: ssm_dt_b_c_rms   = 0
print_info: model type       = 1B
print_info: model params     = 6.94 B
print_info: general.name     = Granite 4.0 H Tiny
print_info: f_embedding_scale = 12.000000
print_info: f_residual_scale  = 0.220000
print_info: f_attention_scale = 0.007813
print_info: n_ff_shexp        = 1024
print_info: vocab type       = BPE
print_info: n_vocab          = 100352
print_info: n_merges         = 100000
print_info: BOS token        = 100257 '<|end_of_text|>'
print_info: EOS token        = 100257 '<|end_of_text|>'
print_info: EOT token        = 100257 '<|end_of_text|>'
print_info: UNK token        = 100269 '<|unk|>'
print_info: PAD token        = 100256 '<|pad|>'
print_info: LF token         = 198 'Ċ'
print_info: FIM PRE token    = 100258 '<|fim_prefix|>'
print_info: FIM SUF token    = 100260 '<|fim_suffix|>'
print_info: FIM MID token    = 100259 '<|fim_middle|>'
print_info: FIM PAD token    = 100261 '<|fim_pad|>'
print_info: EOG token        = 100257 '<|end_of_text|>'
print_info: EOG token        = 100261 '<|fim_pad|>'
print_info: max token length = 256
load_tensors: loading model tensors, this can take a while... (mmap = false)
load_tensors: offloading 10 repeating layers to GPU
load_tensors: offloaded 10/41 layers to GPU
load_tensors:          CPU model buffer size =   120.59 MiB
load_tensors:        CUDA0 model buffer size =  1003.36 MiB
load_tensors:    CUDA_Host model buffer size =  3028.19 MiB
time=2026-03-04T09:37:10.728+01:00 level=INFO source=server.go:1384 msg="waiting for server to become available" status="llm server not responding"
ggml_cuda_host_malloc: failed to allocate 1.00 MiB of pinned memory: out of memory
CUDA error: out of memory
  current device: 0, in function ggml_backend_cuda_device_event_new at C:\a\ollama\ollama\ml\backend\ggml\ggml\src\ggml-cuda\ggml-cuda.cu:4957
  cudaEventCreateWithFlags(&event, 0x02)
C:\a\ollama\ollama\ml\backend\ggml\ggml\src\ggml-cuda\ggml-cuda.cu:94: CUDA error
time=2026-03-04T09:37:23.287+01:00 level=INFO source=server.go:1384 msg="waiting for server to become available" status="llm server loading model"
time=2026-03-04T09:37:24.385+01:00 level=INFO source=server.go:1384 msg="waiting for server to become available" status="llm server not responding"
time=2026-03-04T09:37:26.735+01:00 level=ERROR source=server.go:303 msg="llama runner terminated" error="exit status 1"
time=2026-03-04T09:37:26.902+01:00 level=INFO source=sched.go:518 msg="Load failed" model=C:\Users\user\.ollama\models\blobs\sha256-491ba81786c46a345a5da9a60cdb9f9a3056960c8411dd857153c194b1f91313 error="llama runner process has terminated: CUDA error"
[GIN] 2026/03/04 - 09:37:27 | 500 |   29.7846277s |       127.0.0.1 | POST     "/api/chat"

OS

Windows

GPU

Nvidia

CPU

Intel

Ollama version

0.17.5

Originally created by @ScaryBeats01 on GitHub (Mar 4, 2026). Original GitHub issue: https://github.com/ollama/ollama/issues/14615 ### What is the issue? When I use open-notebook, it allways gives error 500 in just 30 seconds on creating a insight. But if I use the "ollama run" featur, only with commandr-7b:latest gives this errors. I tried qwen3.5:4b, granite3.1-dense:2b, granite4:tiny-h and lfm2.5-thinking without errors. Look: PS C:\Users\user> ollama run command-r7b:latest "Summarize this text in 3 bullet points: >> >> # INPUT >> Neural networks are computing systems inspired by biological neural networks..." Error: 500 Internal Server Error: llama runner process has terminated: CUDA error PS C:\Users\user> ollama run command-r7b:latest "Summarize this text in 3 bullet points: >> >> # INPUT >> Neural networks are computing systems inspired by biological neural networks..."^C PS C:\Users\user> ollama rm command-r7b:latest deleted 'command-r7b:latest' PS C:\Users\user> ollama pull command-r7b:latest pulling manifest pulling b32d935e114c: 100% ▕███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████▏ 5.1 GB pulling 0d8282caa612: 100% ▕███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████▏ 7.2 KB pulling 945eaa8b1428: 100% ▕███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████▏ 13 KB pulling d8455b5dce0b: 100% ▕███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████▏ 110 B pulling 574fdc7616e8: 100% ▕███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████▏ 491 B verifying sha256 digest writing manifest success PS C:\Users\user> ollama run command-r7b:latest "Summarize this text in 3 bullet points: >> >> # INPUT >> Neural networks are computing systems inspired by biological neural networks..." Error: 500 Internal Server Error: llama runner process has terminated: CUDA error PS C:\Users\user> ollama ps NAME ID SIZE PROCESSOR CONTEXT UNTIL ### Relevant log output ```shell print_info: ssm_dt_b_c_rms = 0 print_info: model type = 1B print_info: model params = 6.94 B print_info: general.name = Granite 4.0 H Tiny print_info: f_embedding_scale = 12.000000 print_info: f_residual_scale = 0.220000 print_info: f_attention_scale = 0.007813 print_info: n_ff_shexp = 1024 print_info: vocab type = BPE print_info: n_vocab = 100352 print_info: n_merges = 100000 print_info: BOS token = 100257 '<|end_of_text|>' print_info: EOS token = 100257 '<|end_of_text|>' print_info: EOT token = 100257 '<|end_of_text|>' print_info: UNK token = 100269 '<|unk|>' print_info: PAD token = 100256 '<|pad|>' print_info: LF token = 198 'Ċ' print_info: FIM PRE token = 100258 '<|fim_prefix|>' print_info: FIM SUF token = 100260 '<|fim_suffix|>' print_info: FIM MID token = 100259 '<|fim_middle|>' print_info: FIM PAD token = 100261 '<|fim_pad|>' print_info: EOG token = 100257 '<|end_of_text|>' print_info: EOG token = 100261 '<|fim_pad|>' print_info: max token length = 256 load_tensors: loading model tensors, this can take a while... (mmap = false) load_tensors: offloading 10 repeating layers to GPU load_tensors: offloaded 10/41 layers to GPU load_tensors: CPU model buffer size = 120.59 MiB load_tensors: CUDA0 model buffer size = 1003.36 MiB load_tensors: CUDA_Host model buffer size = 3028.19 MiB time=2026-03-04T09:37:10.728+01:00 level=INFO source=server.go:1384 msg="waiting for server to become available" status="llm server not responding" ggml_cuda_host_malloc: failed to allocate 1.00 MiB of pinned memory: out of memory CUDA error: out of memory current device: 0, in function ggml_backend_cuda_device_event_new at C:\a\ollama\ollama\ml\backend\ggml\ggml\src\ggml-cuda\ggml-cuda.cu:4957 cudaEventCreateWithFlags(&event, 0x02) C:\a\ollama\ollama\ml\backend\ggml\ggml\src\ggml-cuda\ggml-cuda.cu:94: CUDA error time=2026-03-04T09:37:23.287+01:00 level=INFO source=server.go:1384 msg="waiting for server to become available" status="llm server loading model" time=2026-03-04T09:37:24.385+01:00 level=INFO source=server.go:1384 msg="waiting for server to become available" status="llm server not responding" time=2026-03-04T09:37:26.735+01:00 level=ERROR source=server.go:303 msg="llama runner terminated" error="exit status 1" time=2026-03-04T09:37:26.902+01:00 level=INFO source=sched.go:518 msg="Load failed" model=C:\Users\user\.ollama\models\blobs\sha256-491ba81786c46a345a5da9a60cdb9f9a3056960c8411dd857153c194b1f91313 error="llama runner process has terminated: CUDA error" [GIN] 2026/03/04 - 09:37:27 | 500 | 29.7846277s | 127.0.0.1 | POST "/api/chat" ``` ### OS Windows ### GPU Nvidia ### CPU Intel ### Ollama version 0.17.5
GiteaMirror added the bug label 2026-04-22 19:37:01 -05:00
GiteaMirror commented 2026-04-22 19:37:03 -05:00
Author
Owner
Copy Link

@ScaryBeats01 commented on GitHub (Mar 4, 2026):

It occurs in 0.17.6 too.

<!-- gh-comment-id:3996673906 --> @ScaryBeats01 commented on GitHub (Mar 4, 2026): It occurs in 0.17.6 too.
GiteaMirror commented 2026-04-22 19:37:03 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Mar 4, 2026):

CUDA error: out of memory

It helps if you provide the full log. Here are some general methods for dealing with OOMs.

<!-- gh-comment-id:3996798630 --> @rick-github commented on GitHub (Mar 4, 2026): ``` CUDA error: out of memory ``` It helps if you provide the full log. [Here](https://github.com/ollama/ollama/issues/8597#issuecomment-2614533288) are some general methods for dealing with OOMs.
GiteaMirror commented 2026-04-22 19:37:04 -05:00
Author
Owner
Copy Link

@danigixshi commented on GitHub (Mar 4, 2026):

aquí no hay misterio místico ni bug paranormal. El log está cantando la verdad como un karaoke borracho: te estás quedando sin memoria en la GPU. CUDA se queja y el proceso muere, y por eso Ollama te devuelve el Error 500.

La línea clave del log es esta:

ggml_cuda_host_malloc: failed to allocate 1.00 MiB of pinned memory: out of memory

Traducido al castellano humano: la GPU intenta reservar memoria y no queda ni 1 MB libre, así que el runner de llama.cpp se estrella.

Vamos a desmontar lo que pasa.

El modelo command-r7b pesa unos ~5 GB, pero eso no es toda la historia. Los modelos LLM necesitan memoria extra para:

los KV cache (memoria para la conversación)

buffers de cálculo

capas que se cargan en GPU

memoria "pinned" compartida con CPU

Resultado típico para un 7B:

5-6 GB modelo

+2-4 GB buffers

+KV cache

Total real: 8-10 GB de VRAM fácilmente.

Si tu GPU tiene:

6 GB → imposible

8 GB → muy justo

12 GB → debería funcionar

Por eso otros modelos sí funcionan:

qwen3.5:4b

granite tiny

2b

Son mucho más pequeños.

Además en tu log aparece:

offloaded 10/41 layers to GPU
CUDA0 model buffer size = 1003 MB
CUDA_Host buffer = 3028 MB

<!-- gh-comment-id:3999385461 --> @danigixshi commented on GitHub (Mar 4, 2026): aquí no hay misterio místico ni bug paranormal. El log está cantando la verdad como un karaoke borracho: te estás quedando sin memoria en la GPU. CUDA se queja y el proceso muere, y por eso Ollama te devuelve el Error 500. La línea clave del log es esta: ggml_cuda_host_malloc: failed to allocate 1.00 MiB of pinned memory: out of memory Traducido al castellano humano: la GPU intenta reservar memoria y no queda ni 1 MB libre, así que el runner de llama.cpp se estrella. Vamos a desmontar lo que pasa. El modelo command-r7b pesa unos ~5 GB, pero eso no es toda la historia. Los modelos LLM necesitan memoria extra para: los KV cache (memoria para la conversación) buffers de cálculo capas que se cargan en GPU memoria "pinned" compartida con CPU Resultado típico para un 7B: 5-6 GB modelo +2-4 GB buffers +KV cache Total real: 8-10 GB de VRAM fácilmente. Si tu GPU tiene: 6 GB → imposible 8 GB → muy justo 12 GB → debería funcionar Por eso otros modelos sí funcionan: qwen3.5:4b granite tiny 2b Son mucho más pequeños. Además en tu log aparece: offloaded 10/41 layers to GPU CUDA0 model buffer size = 1003 MB CUDA_Host buffer = 3028 MB
GiteaMirror commented 2026-04-22 19:37:04 -05:00
Author
Owner
Copy Link

@ScaryBeats01 commented on GitHub (Mar 7, 2026):

Now works the same command.

<!-- gh-comment-id:4016525063 --> @ScaryBeats01 commented on GitHub (Mar 7, 2026): Now works the same command.
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#35231
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?