github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-07 00:22:43 -05:00
Code Issues 14.5k Packages Projects Releases 120 Wiki Activity

[GH-ISSUE #10969] Request: Allow configurable timeout for “waiting for llama runner to start” #7228

New Issue
Closed
opened 2026-04-12 19:14:23 -05:00 by GiteaMirror · 1 comment
GiteaMirror commented 2026-04-12 19:14:23 -05:00
Owner
Copy Link

Originally created by @brian0913579 on GitHub (Jun 4, 2025).
Original GitHub issue: https://github.com/ollama/ollama/issues/10969

I'm running large models like qwen2.5-coder:32b using Ollama on macOS, with models stored on an external SSD (e.g., /Volumes/T7/ollama-models). When I try to run the model with:

OLLAMA_MODELS=/Volumes/T7/ollama-models ollama run qwen2.5-coder:32b

I consistently get the error:

Error: timed out waiting for llama runner to start - progress 0.00 -

After investigation, it appears this is due to the model taking time to load from the external drive into RAM or GPU memory. However, there is currently no exposed option to extend or configure the timeout duration for the runner startup.

Feature Request

Please allow the timeout to be configurable, either via:

  • an environment variable (e.g., OLLAMA_RUNNER_TIMEOUT)
  • a CLI flag (e.g., --timeout)
  • or a config file setting

Why This Matters

This would be especially useful for users with:

  • large models (13B, 32B)
  • slower or external storage
  • limited memory, which increases load time

Thanks for building such an awesome tool — looking forward to future improvements!

Originally created by @brian0913579 on GitHub (Jun 4, 2025). Original GitHub issue: https://github.com/ollama/ollama/issues/10969 I'm running large models like `qwen2.5-coder:32b` using Ollama on macOS, with models stored on an external SSD (e.g., `/Volumes/T7/ollama-models`). When I try to run the model with: ```bash OLLAMA_MODELS=/Volumes/T7/ollama-models ollama run qwen2.5-coder:32b ``` I consistently get the error: ``` Error: timed out waiting for llama runner to start - progress 0.00 - ``` After investigation, it appears this is due to the model taking time to load from the external drive into RAM or GPU memory. However, there is currently no exposed option to extend or configure the timeout duration for the runner startup. ## Feature Request Please allow the timeout to be configurable, either via: - an environment variable (e.g., `OLLAMA_RUNNER_TIMEOUT`) - a CLI flag (e.g., `--timeout`) - or a config file setting ### Why This Matters This would be especially useful for users with: - large models (13B, 32B) - slower or external storage - limited memory, which increases load time Thanks for building such an awesome tool — looking forward to future improvements!
GiteaMirror added the feature request label 2026-04-12 19:14:23 -05:00
GiteaMirror commented 2026-04-12 19:14:25 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Jun 4, 2025):

OLLAMA_LOAD_TIMEOUT

<!-- gh-comment-id:2939970328 --> @rick-github commented on GitHub (Jun 4, 2025): [`OLLAMA_LOAD_TIMEOUT`](https://github.com/ollama/ollama/blob/5c42800fca4da07d1c362c0f190429993e53c3b5/envconfig/config.go#L261)
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 feature request
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#7228
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?