ent-v2.2.2-base
maximhq/bifrostent-v2.2.2-baseSep 23, 2026by akshaydeo
Upgrading?
AI Summary
Enterprise release introducing pluggable session affinity, typesafe models support, and various provider fixes.
Key Highlights
- Introduced pluggable session affinity with outcome-driven routing
- Added typesafe models support and new decisions endpoint
- Fixed various streaming and provider compatibility issues
New Features
- Pluggable session affinity interface
- Typesafe models support
- New decisions endpoint
- Preamble support for OpenAI model family
- Grok 4.7 reasoning effort preservation
- MCP proxy support fixes
Full Release Notes
## What's Changed * streaming gate fixes by @akshaydeo in https://github.com/maximhq/bifrost/pull/7331 * stream gate memory fixes by @akshaydeo in https://github.com/maximhq/bifrost/pull/7332 * adds more gating test cases by @akshaydeo in https://github.com/maximhq/bifrost/pull/7333 * enterprise docs by @akshaydeo in https://github.com/maximhq/bifrost/pull/7334 * enterprise docs by @akshaydeo in https://github.com/maximhq/bifrost/pull/7335 * feat: extract session stickiness into a pluggable `SessionAffinity` interface with scoped, identity-aware key bindings by @Pratham-Mishra04 in https://github.com/maximhq/bifrost/pull/7176 * feat: route and key session affinity driven by observed outcomes, with `x-bf-session-affinity` header opt-out by @Pratham-Mishra04 in https://github.com/maximhq/bifrost/pull/7177 * feat: add `BifrostContextKeySessionAffinity` for per-request session affinity control and document provider-level session binding behavior by @Pratham-Mishra04 in https://github.com/maximhq/bifrost/pull/7178 * refactor: export `SessionStateKey`, `SessionAffinity`, and related helpers; unify complexity session keying under shared session state by @Pratham-Mishra04 in https://github.com/maximhq/bifrost/pull/7179 * docs: add Session Affinity guide and cross-link it from routing, CLI agent, and request-options docs by @Pratham-Mishra04 in https://github.com/maximhq/bifrost/pull/7210 * fix: record session affinity agreement with routing in trail and engines-used, log unweighted provider exclusions, clarify docs by @Pratham-Mishra04 in https://github.com/maximhq/bifrost/pull/7289 * [fix]: preserve API web search source names by @g-yixuan in https://github.com/maximhq/bifrost/pull/7358 * fix(anthropic): carry allowed_tools and parallel_tool_calls into the request by @basil-k-aji-dev in https://github.com/maximhq/bifrost/pull/7348 * fix(bedrock): keep cache_control when a message's content is a string by @basil-k-aji-dev in https://github.com/maximhq/bifrost/pull/7360 * dependabot cleanups by @akshaydeo in https://github.com/maximhq/bifrost/pull/7378 * fix: allow all providers access and api models response by @TejasGhatte in https://github.com/maximhq/bifrost/pull/7375 * typesafe models support, new decisions endpoint by @akshaydeo in https://github.com/maximhq/bifrost/pull/7355 * feat(decision): emulate decisions via LLM tool-calling for all providers by @akshaydeo in https://github.com/maximhq/bifrost/pull/7361 * benchmarking and api struct fixes by @akshaydeo in https://github.com/maximhq/bifrost/pull/7384 * fix: removes unnecessary allocations from compat plugin by @sammaji in https://github.com/maximhq/bifrost/pull/7097 * fix: strip bedrock cachepoint markers for non-bedrock providers while preserving them for fallbacks by @sammaji in https://github.com/maximhq/bifrost/pull/7182 * adds bedrock unknown json document handling by @akshaydeo in https://github.com/maximhq/bifrost/pull/7396 * [fix]: preserve Grok 4.7 xhigh reasoning effort by @nettee in https://github.com/maximhq/bifrost/pull/7403 * docs: add prompt deployment docs by @roroghost17 in https://github.com/maximhq/bifrost/pull/7420 * adds safeguard support for passthrough requests by @akshaydeo in https://github.com/maximhq/bifrost/pull/7393 * memory fixes for streaming cancellation by @akshaydeo in https://github.com/maximhq/bifrost/pull/7406 * fix(responses): stop rewriting OpenAI computer tool to computer_use_preview by @abhishekgahlot2 in https://github.com/maximhq/bifrost/pull/7426 * perf: avoid redundant JSON decoding in Responses requests by @G-XD in https://github.com/maximhq/bifrost/pull/7412 * fix(schemas): deep copy extended Responses fields in DeepCopyResponsesMessage by @Madhuvod in https://github.com/maximhq/bifrost/pull/7422 * ci: add CodSpeed performance benchmarks for the request hot path by @codspeed[bot] in https://github.com/maximhq/bifrost/pull/7409 * readact anthropic billing headers to avoid cache spoiling by @akshaydeo in https://github.com/maximhq/bifrost/pull/7431 * kimi and deepseek support for claude code by @akshaydeo in https://github.com/maximhq/bifrost/pull/7430 * fix: claude opus 5-5 changes by @TejasGhatte in https://github.com/maximhq/bifrost/pull/7433 * fix: restore `HTTP_PROXY`/`HTTPS_PROXY` support for MCP connections while blocking link-local targets via pre-proxy destination guard by @Pratham-Mishra04 in https://github.com/maximhq/bifrost/pull/7437 * test fixes by @akshaydeo in https://github.com/maximhq/bifrost/pull/7440 * fix: computer toolset compatibility by @TejasGhatte in https://github.com/maximhq/bifrost/pull/7434 * fix: computer toolset model capability by @TejasGhatte in https://github.com/maximhq/bifrost/pull/7441 * v2.2.2 by @akshaydeo in https://github.com/maximhq/bifrost/pull/7442 * changelogs for 2.2.2 by @akshaydeo in https://github.com/maximhq/bifrost/pull/7443 * adds preamble support for all openai model family by @akshaydeo in https://github.com/maximhq/bifrost/pull/7444 * preamble support for openai model family across the providers by @akshaydeo in https://github.com/maximhq/bifrost/pull/7445 * docs fixes by @akshaydeo in https://github.com/maximhq/bifrost/pull/7447 ## New Contributors * @g-yixuan made their first contribution in https://github.com/maximhq/bifrost/pull/7358 * @basil-k-aji-dev made their first contribution in https://github.com/maximhq/bifrost/pull/7348 * @abhishekgahlot2 made their first contribution in https://github.com/maximhq/bifrost/pull/7426 * @codspeed[bot] made their first contribution in https://github.com/maximhq/bifrost/pull/7409 **Full Changelog**: https://github.com/maximhq/bifrost/compare/ent-v2.2.1-base...ent-v2.2.2-base
Related Repositories
ollama/ollama★ 184,214
Get up and running with Kimi, GLM, MiniMax, DeepSeek, gpt-oss, Qwen, Gemma and other models.
fatedier/frp★ 110,053
A fast reverse proxy to help you expose a local server behind a NAT or firewall to the internet.
infiniflow/ragflow★ 92,614
RAGFlow is a leading open-source Retrieval-Augmented Generation (RAG) engine that fuses cutting-edge RAG with Agent capabilities to create a superior context layer for LLMs
gohugoio/hugo★ 90,198
The world’s fastest framework for building websites.
junegunn/fzf★ 83,716
:cherry_blossom: A command-line fuzzy finder