jianf/ModelRouter
1
0
Fork 0
mirror of https://gitcode.com/JianFeeeee/ModelRouter.git synced 2026-09-20 00:48:00 +00:00
Code Issues Packages Projects Releases Wiki Activity

feat(gateway): pass through upstream finish_reason end-to-end

Browse Source 
The gateway hardcoded "stop" on every terminating stream chunk, so
tool-call rounds reported finish_reason=stop and length caps were
invisible to clients. UnifiedChunk now carries finish_reason; adapters
emit it (with empty-string finish reasons like sensenova treated as
non-terminal), standardSSEChunk passes it through for un-adapted
upstreams, [DONE] no longer emits a duplicate reason-less done chunk,
and both streaming paths emit the real reason with "stop" as fallback.

Also vendor sensenova/agentrouter adapters into the repo: they were
WebUI-only uploads and a deploy sync silently removed them while live
AUTO-chain slots still referenced them.
This commit is contained in:
JianFeeeee
2026-08-24 15:05:50 +08:00
parent 9c99ded8c0
commit bb3af3bdb3
17 changed files with 399 additions and 27 deletions
Download Patch File Download Diff File Expand all files Collapse all files

32
internal/gateway/chat.go
View File

@ -656,6 +656,7 @@ func (g *Gateway) streamChat(w http.ResponseWriter, ctx context.Context, cands [
return
}
var lastUsage *types.TokenUsage
lastFinish := ""
for ck := range chunks {
if ck.Usage != nil {
lastUsage = mergeUsage(lastUsage, ck.Usage)
@ -672,8 +673,12 @@ func (g *Gateway) streamChat(w http.ResponseWriter, ctx context.Context, cands [
}
choice := ChunkChoice{Index: 0, Delta: delta}
if ck.Done {
stop := "stop"
choice.FinishReason = &stop
fin := ck.FinishReason
if fin == "" {
fin = "stop"
}
lastFinish = fin
choice.FinishReason = &fin
}
chunk.Choices = []ChunkChoice{choice}
rec.Compl += int64(len(ck.Content)+len(ck.ReasoningContent)+len(ck.ToolCalls)) / 3
@ -681,10 +686,13 @@ func (g *Gateway) streamChat(w http.ResponseWriter, ctx context.Context, cands [
return
}
}
stop := "stop"
finalFinish := lastFinish
if finalFinish == "" {
finalFinish = "stop"
}
send(ChatChunk{
ID: id, Object: "chat.completion.chunk", Created: created, Model: effective,
Choices: []ChunkChoice{{Index: 0, Delta: RespMessage{}, FinishReason: &stop}},
Choices: []ChunkChoice{{Index: 0, Delta: RespMessage{}, FinishReason: &finalFinish}},
})
// Final usage chunk (OpenAI standard: empty choices + usage before [DONE]).
// Prefer the upstream's exact usage if the stream carried it; fall back to
@ -826,6 +834,7 @@ func (g *Gateway) streamChatAuto(w http.ResponseWriter, ctx context.Context, cha
return
}
var lastUsage *types.TokenUsage
lastFinish := ""
for ck := range chunks {
if ck.Usage != nil {
lastUsage = mergeUsage(lastUsage, ck.Usage)
@ -842,8 +851,12 @@ func (g *Gateway) streamChatAuto(w http.ResponseWriter, ctx context.Context, cha
}
choice := ChunkChoice{Index: 0, Delta: delta}
if ck.Done {
stop := "stop"
choice.FinishReason = &stop
fin := ck.FinishReason
if fin == "" {
fin = "stop"
}
lastFinish = fin
choice.FinishReason = &fin
}
chunk.Choices = []ChunkChoice{choice}
rec.Compl += int64(len(ck.Content)+len(ck.ReasoningContent)+len(ck.ToolCalls)) / 3
@ -851,10 +864,13 @@ func (g *Gateway) streamChatAuto(w http.ResponseWriter, ctx context.Context, cha
return
}
}
stop := "stop"
finalFinish := lastFinish
if finalFinish == "" {
finalFinish = "stop"
}
send(ChatChunk{
ID: id, Object: "chat.completion.chunk", Created: created, Model: rec.Model,
Choices: []ChunkChoice{{Index: 0, Delta: RespMessage{}, FinishReason: &stop}},
Choices: []ChunkChoice{{Index: 0, Delta: RespMessage{}, FinishReason: &finalFinish}},
})
// Final usage chunk (OpenAI standard: empty choices + usage before [DONE]).
// Prefer the upstream's exact usage if the stream carried it; fall back to