luulam/9router
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

fix(minimax): preserve images on matched OpenAI transport

Browse Source 
Prefer the sourceFormat-matched runtime transport over a model's
declared targetFormat when both apply. MiniMax-M3 previously resolved
to a Claude-shaped body while being posted to the already-selected
OpenAI endpoint, silently dropping image_url blocks from OpenAI
clients. Fixes #3418.
This commit is contained in:
turingcat
2026-08-28 16:33:29 +07:00
parent 2a9213c5bd
commit 28d005772a
2 changed files with 195 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

7
open-sse/handlers/chatCore.js
View File

@@ -91,7 +91,12 @@ export async function handleChatCore({ body, modelInfo, credentials, log, onCred
// differ — kimi/glm only do /chat/completions). Undeclared models keep the
// upstream default (use the transport), preserving behavior for glm/deepseek/...
const useTransport = (!modelSupportedFormats || modelSupportedFormats.includes(sourceFormat)) ? runtimeTransport : null;
const targetFormat = modelTargetFormat || useTransport?.format || getTargetFormat(provider, credentials);
// A source-format-matched endpoint keeps the request lossless. Prefer it
// over a model-level targetFormat, which is only the fallback for clients
// whose wire format has no supported transport (for example MiniMax-M3:
// OpenAI clients should stay on /chat/completions; other clients can fall
// back to its declared Claude target).
const targetFormat = useTransport?.format || modelTargetFormat || getTargetFormat(provider, credentials);
if (useTransport && credentials) credentials.runtimeTransport = useTransport;
const stripList = getModelStrip(alias, model);
const upstreamModel = getModelUpstreamId(alias, model);