github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-21 13:11:39 -05:00
Code Issues 16.9k Packages Projects Releases 122 Wiki Activity

[GH-ISSUE #920] [Starcoder:7b] Not using CUDA #78125

New Issue
Closed
opened 2026-05-08 22:01:01 -05:00 by GiteaMirror · 2 comments
GiteaMirror commented 2026-05-08 22:01:01 -05:00
Owner
Copy Link

Originally created by @ManuLinares on GitHub (Oct 26, 2023).
Original GitHub issue: https://github.com/ollama/ollama/issues/920

Originally assigned to: @BruceMacD on GitHub.

# ollama run starcoder:7b prompt

...
llm_load_tensors: VRAM used: 3968.42 MB
..GGML_ASSERT: /build/ollama/src/ollama-cuda/llm/llama.cpp/gguf/ggml-cuda.cu:6115: false
2023/10/26 16:09:13 llama.go:378: signal: aborted (core dumped)
2023/10/26 16:09:13 llama.go:386: error starting llama runner: llama runner process has terminated
2023/10/26 16:09:13 llama.go:452: llama runner stopped successfully
2023/10/26 16:09:13 llama.go:363: starting llama runner
2023/10/26 16:09:13 llama.go:421: waiting for llama runner to start responding
{"timestamp":1698347353,"level":"WARNING","function":"server_params_parse","line":871,"message":"Not compiled with GPU offload support, --n-gpu-layers option will be ignored. See main README.md for information on enabling GPU BLAS support","n_gpu_layers":-1}
...

Other models work fine:

example with ollama run mistral:instruct:

...
2023/10/26 16:12:55 llama.go:252: 7456 MB VRAM available, loading up to 55 GPU layers
2023/10/26 16:12:55 llama.go:363: starting llama runner
2023/10/26 16:12:55 llama.go:421: waiting for llama runner to start responding
ggml_init_cublas: found 1 CUDA devices:
  Device 0: NVIDIA GeForce RTX 3070, compute capability 8.6
...
Originally created by @ManuLinares on GitHub (Oct 26, 2023). Original GitHub issue: https://github.com/ollama/ollama/issues/920 Originally assigned to: @BruceMacD on GitHub. `# ollama run starcoder:7b prompt` ``` ... llm_load_tensors: VRAM used: 3968.42 MB ..GGML_ASSERT: /build/ollama/src/ollama-cuda/llm/llama.cpp/gguf/ggml-cuda.cu:6115: false 2023/10/26 16:09:13 llama.go:378: signal: aborted (core dumped) 2023/10/26 16:09:13 llama.go:386: error starting llama runner: llama runner process has terminated 2023/10/26 16:09:13 llama.go:452: llama runner stopped successfully 2023/10/26 16:09:13 llama.go:363: starting llama runner 2023/10/26 16:09:13 llama.go:421: waiting for llama runner to start responding {"timestamp":1698347353,"level":"WARNING","function":"server_params_parse","line":871,"message":"Not compiled with GPU offload support, --n-gpu-layers option will be ignored. See main README.md for information on enabling GPU BLAS support","n_gpu_layers":-1} ... ``` Other models work fine: example with **ollama run mistral:instruct**: ``` ... 2023/10/26 16:12:55 llama.go:252: 7456 MB VRAM available, loading up to 55 GPU layers 2023/10/26 16:12:55 llama.go:363: starting llama runner 2023/10/26 16:12:55 llama.go:421: waiting for llama runner to start responding ggml_init_cublas: found 1 CUDA devices: Device 0: NVIDIA GeForce RTX 3070, compute capability 8.6 ... ```
GiteaMirror added the bug label 2026-05-08 22:01:01 -05:00
GiteaMirror commented 2026-05-08 22:01:05 -05:00
Author
Owner
Copy Link

@rhokstar commented on GitHub (Oct 30, 2023):

@ManuLinares @jmorganca Thanks for reporting! This bug was driving me crazy! I have the same problem as well. I'm running 2 GPUs: 1080 GTX and RTX A6000. When I prompt Star Coder, my CPU is being used. But when I run Mistral, my A6000 is working (I specified this through nvidia-smi).

<!-- gh-comment-id:1785415814 --> @rhokstar commented on GitHub (Oct 30, 2023): @ManuLinares @jmorganca Thanks for reporting! This bug was driving me crazy! I have the same problem as well. I'm running 2 GPUs: 1080 GTX and RTX A6000. When I prompt Star Coder, my CPU is being used. But when I run Mistral, my A6000 is working (I specified this through nvidia-smi).
GiteaMirror commented 2026-05-08 22:01:06 -05:00
Author
Owner
Copy Link

@mxyng commented on GitHub (Nov 21, 2023):

Should be fixed with #1224

<!-- gh-comment-id:1821635001 --> @mxyng commented on GitHub (Nov 21, 2023): Should be fixed with #1224
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#78125
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?