fix: add opencode-zen, xiaomi, nvidia to STREAMING_ENDPOINT_PROVIDERS#2141
Open
TF0rd wants to merge 1 commit into
Open
fix: add opencode-zen, xiaomi, nvidia to STREAMING_ENDPOINT_PROVIDERS#2141TF0rd wants to merge 1 commit into
TF0rd wants to merge 1 commit into
Conversation
The `response-mode-guard` silently drops routes from providers not in
`STREAMING_ENDPOINT_PROVIDERS` when the tier's response mode is
'stream'. This causes complex-tier fallback chains to skip
opencode-zen (MiniMax M3) and xiaomi (MiMo-V2.5-Pro) entirely,
promoting the second or third fallback to primary without any log
output or user-visible indication.
All three providers expose OpenAI-compatible SSE streaming:
- opencode-zen: opencode.ai/zen/v1/chat/completions
- xiaomi: token-plan-{region}.xiaomimimo.com/v1/chat/completions
- nvidia: integrate.api.nvidia.com/v1/chat/completions
Their PROVISIONER_ENDPOINTS entries already include
`...openaiStreamUsage`, confirming server-side streaming support.
![cubic-dev-ai[bot]](https://avatars.githubusercontent.com/in/1082092?s=60&v=4){kind=small}
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Problem
The
response-mode-guardsilently drops routes from providers not inSTREAMING_ENDPOINT_PROVIDERSwhen a tier usesresponse_mode: stream. For users routing through opencode-zen or xiaomi, this means their primary and fallback models are filtered out at runtime with no warning, and the first stream-capable fallback gets silently promoted — which is confusing and defeats the purpose of having fallback chains.Specifically:
All three providers support SSE streaming. Their entries in
PROVIDER_ENDPOINTS(provider-endpoints.ts) already include...openaiStreamUsage, which configures the proxy to requeststream_options.include_usagefor streaming responses.Fix
Add
'opencode-zen','xiaomi', and'nvidia'to theSTREAMING_ENDPOINT_PROVIDERSset inmodel-capabilities.ts.Related
PROVIDER_ENDPOINTSentries for xiaomi-subscription (L246-250), opencode-zen (L419-424), and nvidia (L267-272) inprovider-endpoints.tsSummary by cubic
Add
opencode-zen,xiaomi, andnvidiatoSTREAMING_ENDPOINT_PROVIDERSso streaming tiers route correctly instead of being dropped byresponse-mode-guard. All three support SSE via OpenAI‑compatible chat completions; no other changes.Written for commit 14eb380. Summary will update on new commits.