ent-v2.2.5-base
maximhq/bifrostent-v2.2.5-baseOct 2, 2026by akshaydeo
Upgrading?
AI Summary
Enterprise base version 2.2.5 introduces a security feature for initial setup and fixes billing and authentication logic.
Key Highlights
- Added first-time setup token for security
- Fixed billing for ultrafast requests exceeding 272k tokens
- Implemented auth path whitelisting
New Features
- First-time setup token
Full Release Notes
## What's Changed * fix(guardrails): align raw transform target IDs with normalized guardrail ordinals for billing header and MCP blocks by @Madhuvod in https://github.com/maximhq/bifrost/pull/7808 * fix: bill ultrafast requests above 272k tokens at the long-context ultrafast rate by @impoiler in https://github.com/maximhq/bifrost/pull/7834 * adds first time setup token to avoid opening new setup to the world (#5759) by @akshaydeo in https://github.com/maximhq/bifrost/pull/7830 * path normalization auth bypass (#5763) by @akshaydeo in https://github.com/maximhq/bifrost/pull/7831 * auth path whitelisting by @akshaydeo in https://github.com/maximhq/bifrost/pull/7832 * enforce autoexecute allow list by @akshaydeo in https://github.com/maximhq/bifrost/pull/7833 * gemini fallback to openai fixes by @akshaydeo in https://github.com/maximhq/bifrost/pull/7835 * fix: bill openai service_tier fast at the priority rates and echo it to clients by @akshaydeo in https://github.com/maximhq/bifrost/pull/7837 * feat: add the priority above-272k cache write pricing column so fast long-context cache writes bill at the published rate by @akshaydeo in https://github.com/maximhq/bifrost/pull/7838 * test fixes by @akshaydeo in https://github.com/maximhq/bifrost/pull/7839 * 2.2.5 release notes by @akshaydeo in https://github.com/maximhq/bifrost/pull/7840 * test fixes by @akshaydeo in https://github.com/maximhq/bifrost/pull/7845 **Full Changelog**: https://github.com/maximhq/bifrost/compare/ent-v2.2.4-base...ent-v2.2.5-base
Related Repositories
ollama/ollama★ 184,258
Get up and running with Kimi, GLM, MiniMax, DeepSeek, gpt-oss, Qwen, Gemma and other models.
fatedier/frp★ 110,072
A fast reverse proxy to help you expose a local server behind a NAT or firewall to the internet.
infiniflow/ragflow★ 92,646
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★ 90,197
The world’s fastest framework for building websites.
junegunn/fzf★ 83,716
:cherry_blossom: A command-line fuzzy finder