Skip to content

Use Claude Code with your ChatGPT, Kimi, Cursor or Grok subscription via a local Anthropic-compatible proxy

Rust

remove unsupported max_output_tokens

Raine Virta350bb3114f05
diff --git a/src/server.ts b/src/server.ts
index b9e1cd8..90e79a1 100644
--- a/src/server.ts
+++ b/src/server.ts
@@ -218,7 +218,6 @@ async function handleMessages(req: Request, reqId: string): Promise<Response> {
tools: translated.tools?.length ?? 0,
hasInstructions: !!translated.instructions,
requestedMaxTokens: body.max_tokens,
- translatedMaxOutputTokens: translated.max_output_tokens,
hasContextManagement: contextManagement !== undefined,
promptCacheKey: translated.prompt_cache_key,
})
@@ -239,7 +238,6 @@ async function handleMessages(req: Request, reqId: string): Promise<Response> {
translatedToolCount: translated.tools?.length ?? 0,
hasInstructions: !!translated.instructions,
requestedMaxTokens: body.max_tokens,
- translatedMaxOutputTokens: translated.max_output_tokens,
hasContextManagement: contextManagement !== undefined,
contextManagement,
previousCountReqId: state?.lastCount?.reqId,
@@ -296,7 +294,6 @@ async function handleMessages(req: Request, reqId: string): Promise<Response> {
localInputTokens,
translatedInputTokens,
requestedMaxTokens: body.max_tokens,
- translatedMaxOutputTokens: translated.max_output_tokens,
hasContextManagement: contextManagement !== undefined,
contextManagement,
upstreamInputTokens: finish.usage?.input_tokens ?? 0,
@@ -339,7 +336,6 @@ async function handleMessages(req: Request, reqId: string): Promise<Response> {
localInputTokens,
translatedInputTokens,
requestedMaxTokens: body.max_tokens,
- translatedMaxOutputTokens: translated.max_output_tokens,
hasContextManagement: contextManagement !== undefined,
contextManagement,
upstreamInputTokens: result.rawUsage?.input_tokens ?? 0,
diff --git a/src/translate/request.ts b/src/translate/request.ts
index 8d2b50f..ec5a393 100644
--- a/src/translate/request.ts
+++ b/src/translate/request.ts
@@ -12,7 +12,6 @@ export interface ResponsesRequest {
instructions?: string
input: ResponsesInputItem[]
tools?: ResponsesTool[]
- max_output_tokens?: number
tool_choice?:
| "auto"
| "none"
@@ -96,7 +95,6 @@ export function translateRequest(req: AnthropicRequest, opts: TranslateOptions =
}
if (instructions) out.instructions = instructions
if (tools && tools.length) out.tools = tools
- if (req.max_tokens) out.max_output_tokens = req.max_tokens
if (opts.sessionId) out.prompt_cache_key = opts.sessionId
const effort = req.output_config?.effort
if (effort) out.reasoning = { effort }