Issues / #545

#545 Error: 400: {"type":"invalid_request_error","message":"prompt (38963 tokens) + max tokens (226281) exceeds the context (262144); requests are never truncated"}

closed · @avarakin · 4 commentaires · Sur GitHub

Models & quants

Description

Getting stuck with this error even after compacting. Increasing context does not help. I am using Pi.dev 1.0 harness. This issue does not happen with qwen 3.8 27b. Machine has 16g VRAM and 64g RAM.

Sur le site

Liens install, modèles, releases.