feat(bixarena): implement backend streaming for LLM chat completions (SMR-698)#3931

Draft

rrchai wants to merge 20 commits intoSage-Bionetworks:mainfrom

rrchai:bixarena/llm-ai-service

Commits on Mar 20, 2026

feat(bixarena): add OpenAPI schemas for ai-service POST /chat/completions
rrchai
committed
refactor(bixarena): simplify chat schemas by reusing existing Message and ModelError schemas
rrchai
committed
build(bixarena): generate ai-service code for POST /chat/completions
rrchai
committed
feat(bixarena): add apiModelName and apiBase to ModelChatRequest schema
rrchai
committed
build(bixarena): bump FastAPI to >=0.135.0 for native SSE support
rrchai
committed
feat(bixarena): implement ai-service SSE chat completion endpoint
rrchai
committed
feat(bixarena): add chat completion generation params and timeout config
rrchai
committed
feat(bixarena): add app attribution headers for provider analytics
rrchai
committed
feat(bixarena): add OpenAPI spec for round stream endpoint
rrchai
committed
feat(bixarena): implement api-service SSE round stream proxy endpoint
rrchai
committed
fix(bixarena): fix ai-service SSE output format
rrchai
committed
feat(bixarena): register chat completion route in ai-service app
rrchai
committed
feat(bixarena): hardcode OpenRouter app attribution defaults
rrchai
committed
feat(bixarena): add application logging config to ai-service
rrchai
committed
fix(bixarena): fix SSE usage data capture in streaming chunks
rrchai
committed
build(bixarena): regenerate gateway routes with stream endpoint
rrchai
committed
docs(bixarena): add LOG_LEVEL to ai-service .env.example
rrchai
committed
build(bixarena): regenerate all clients and gateway routes
rrchai
committed
build(bixarena): regenerate all clients after rebase onto main
rrchai
committed
chore: revert settings.json to upstream/main
rrchai
committed

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

feat(bixarena): implement backend streaming for LLM chat completions (SMR-698)#3931

feat(bixarena): implement backend streaming for LLM chat completions (SMR-698)#3931
rrchai wants to merge 20 commits intoSage-Bionetworks:mainfrom
rrchai:bixarena/llm-ai-service

Commits on Mar 20, 2026

feat(bixarena): add OpenAPI schemas for ai-service POST /chat/completions

refactor(bixarena): simplify chat schemas by reusing existing Message and ModelError schemas

build(bixarena): generate ai-service code for POST /chat/completions

feat(bixarena): add apiModelName and apiBase to ModelChatRequest schema

build(bixarena): bump FastAPI to >=0.135.0 for native SSE support

feat(bixarena): implement ai-service SSE chat completion endpoint

feat(bixarena): add chat completion generation params and timeout config

feat(bixarena): add app attribution headers for provider analytics

feat(bixarena): add OpenAPI spec for round stream endpoint

feat(bixarena): implement api-service SSE round stream proxy endpoint

fix(bixarena): fix ai-service SSE output format

feat(bixarena): register chat completion route in ai-service app

feat(bixarena): hardcode OpenRouter app attribution defaults

feat(bixarena): add application logging config to ai-service

fix(bixarena): fix SSE usage data capture in streaming chunks

build(bixarena): regenerate gateway routes with stream endpoint

docs(bixarena): add LOG_LEVEL to ai-service .env.example

build(bixarena): regenerate all clients and gateway routes

build(bixarena): regenerate all clients after rebase onto main

chore: revert settings.json to upstream/main