Patch 4: Increase OpenRouter promptWindowLimit fallback via AGENT_CONTEXT_WINDOW_FALLBACK
Both static and instance fallbacks increased from 4096 to 2,000,000 tokens. When model isn't in models.json cache (e.g. Grok), it was falling back to 4096 tokens causing severe truncation of file reads. Reuses AGENT_CONTEXT_WINDOW_FALLBACK env var from Patch 2.
This commit is contained in:
parent
5009ddbb3a
commit
5bcef7d604
@ -168,7 +168,7 @@ class OpenRouterLLM {
|
||||
{}
|
||||
)
|
||||
: {};
|
||||
return availableModels[modelName]?.maxLength || 4096;
|
||||
return availableModels[modelName]?.maxLength || Number(process.env.AGENT_CONTEXT_WINDOW_FALLBACK) || 2_000_000;
|
||||
}
|
||||
|
||||
promptWindowLimit() {
|
||||
|
||||
Loading…
Reference in New Issue
Block a user