Ollama llama runner cgo crash — exit status 2 on specific GPU/CUDA combos

What this error means

llama runner process has terminated: signal arrived during cgo execution is a Ollama failure pattern reported for developers trying to fix ollama llama runner crash with ‘signal arrived during cgo execution’ on nvidia gpu. Based on the imported evidence, treat this as a tool-specific troubleshooting page rather than a generic API error.

Why this happens

Fresh GitHub issue (#16102, created 2026-05-11) on ollama/ollama. GTX TITAN X (12 GB VRAM) + CUDA 13 + Ubuntu crashes on Qwen2.5 14B with ‘llama runner process has terminated: signal arrived during cgo execution’. VRAM only reaches ~1 GB before crash, ruling out OOM. 7B models load fine on same hardware.

Common causes

• Ollama users with older NVIDIA GPUs (Maxwell architecture) and CUDA 13 drivers experience immediate crashes when loading Qwen2.5 14B models. The 7B models work fine, suggesting a backend compatibility issue with specific model sizes. Error surfaces as HTTP 500 with ‘exit status 2’ in Ollama logs.
• Fresh GitHub issue (#16102, created 2026-05-11) on ollama/ollama. GTX TITAN X (12 GB VRAM) + CUDA 13 + Ubuntu crashes on Qwen2.5 14B with ‘llama runner process has terminated: signal arrived during cgo execution’. VRAM only reaches ~1 GB before crash, ruling out OOM. 7B models load fine on same hardware.

Quick fixes

1. Confirm the exact error signature matches llama runner process has terminated: signal arrived during cgo execution.
2. Check the Ollama account, local tool state, and provider configuration involved in the failing workflow.
3. Compare the failing environment with a known working setup, then change one configuration value at a time.

Platform/tool-specific checks

• Verify the command, editor, extension, or API client that produced the error.
• Compare local settings with CI, deployment, or editor-level settings when the error appears in only one environment.
• Avoid deleting credentials, local model data, or project settings until the failing scope is clear.

Step-by-step troubleshooting

1. Capture the exact error message and the command, editor action, or request that triggered it.
2. Check whether the failure is account/auth, quota/rate, model/provider, local runtime, or deployment configuration.
3. Review the source evidence below and compare it with your environment.
4. Apply one change at a time and rerun the smallest failing action.
5. Keep the working fix documented for the team or deployment environment.

How to prevent it

• Keep provider/tool configuration documented.
• Record non-secret diagnostics such as tool version, provider name, model name, and command path.
• Add a lightweight check before CI or production workflows depend on the tool.

Sources checked

• https://github.com/ollama/ollama/issues/16102

Evidence note: Fresh GitHub issue (#16102, created 2026-05-11) on ollama/ollama. GTX TITAN X (12 GB VRAM) + CUDA 13 + Ubuntu crashes on Qwen2.5 14B with ‘llama runner process has terminated: signal arrived during cgo execution’. VRAM only reaches ~1 GB before crash, ruling out OOM. 7B models load fine on same hardware.

Related errors

• Ollama GPU memory allocation failed
• Ollama CUDA driver incompatibility
• Ollama model pull fails with connection refused

FAQ

What should I check first?

Start with the exact llama runner process has terminated: signal arrived during cgo execution text and the smallest action that reproduces it.

Can I ignore this error?

No. Treat it as a failed Ollama workflow until the root cause is understood.

Is this guaranteed to have one fix?

No. The imported evidence supports the troubleshooting path above, but tool behavior can vary by account, plan, version, provider, and local configuration.

How do I know the fix worked?

Rerun the same command, editor action, or request. The fix is working when that action completes without llama runner process has terminated: signal arrived during cgo execution.