Model pricing fetch failed | Friends of the Crustacean 🦞🤝 | Page 1

thorn needle Apr 28, 2026, 4:26 AM

#

On every cold boot, [model-pricing] logs a 60s timeout:
[model-pricing] LiteLLM pricing fetch failed (timeout 60s):
TimeoutError: The operation was aborted due to timeout
The container can reach litellm.ai directly (curl returns HTTP redirect in <5s), so it's not a network/DNS issue.
Running v2026.4.26. Questions:

What endpoint does the pricing fetch actually hit — is it the LiteLLM public API or something else?
Is there a config option to disable the pricing fetch entirely, or extend/reduce the timeout?
Is this fetch blocking anything functional, or is it purely informational at boot?

crimson spadeBOT Apr 28, 2026, 4:26 AM

#

To help others find answers, you can mark your question as solved via Right click solution message -> Apps -> ✅ Mark Solution

barren wrenBOT Apr 28, 2026, 4:27 AM

#

Checked v2026.4.26 source.

LiteLLM pricing endpoint: it does not hit litellm.ai. It fetches the raw GitHub catalog:

https://raw.githubusercontent.com/BerriAI/litellm/main/model_prices_and_context_window.json

It also fetches OpenRouter pricing in parallel from:

https://openrouter.ai/api/v1/models

So I’d test the exact URL from the container:

#Model pricing fetch failed