BerriAI/litellm v1.77.1-nightly

on GitHub

What's Changed

• Revert "Add additionalProperties to vertex ai Schema definition" by @krrishdholakia in #14466
• [Feat] Cost Tracking - Add support for Tiered Cost Tracking for Qwen API (Dashscope) by @ishaan-jaff in #14471
• [Bug]: Azure OpenAI & AI Foundry Reject Image Generation Payload Due to extra_body Injection in LiteLLM v1.76.3 by @ishaan-jaff in #14475
• [Fixes] Using Qwen API Tiered Pricing by @ishaan-jaff in #14479
• fix vertex ai file upload by @Sameerlite in #14460
• Fix provider budgets by @holzman in #14459
• propagate execution context into logging tasks by @dinmukhamedm in #14455
• User Headers X LiteLLM Users Mapping feature by @boopesh07 in #14485
• Litellm gemini cli bug fix by @kankute-sameer in #14451
• Fix gemini cli error by @Sameerlite in #14417
• Revert "fix vertex ai file upload" by @ishaan-jaff in #14501
• [Feat] Bedrock Batches - Working e2e flow to upload file + create batch by @ishaan-jaff in #14518
• [Performance] Use _PROXY_MaxParallelRequestsHandler_v3 by default again by @Bobronium in #14450
• [Feat] Bedrock Batches - Ensure correct transformation applied to incoming requests by @ishaan-jaff in #14522
• [Docs] - Bedrock Batches API by @ishaan-jaff in #14526
• [QA] E2E - Testing for bedrock batches api by @ishaan-jaff in #14525
• Fixed Log Tab Key Alias filtering inaccurately for failed logs by @sashank5644 in #14469
• /v1/messages - don't send content block after message w/ finish reason + usage block + /key/unblock - support hashed tokens by @krrishdholakia in #14477
• Feature - new litellm_request_debug=true flag, enables emitting raw request/response log on single request by @krrishdholakia in #14482
• Fixed Log Tab Key Alias by @krrishdholakia in #14529
• (feat) Model Group Header Forwarding - support wildcard models + add docs by @krrishdholakia in #14528
• fix(lm_studio): resolve illegal Bearer header value issue by @timelfrink in #14512
• Noma non blocking monitor mode & anonymize input support by @TomAlon in #14401
• build(deps): bump hono from 4.6.5 to 4.9.7 in /litellm-js/spend-logs by @dependabot[bot] in #14513
• fix: remove dynamic creation of static value by @AlexsanderHamir in #14538

New Contributors

• @Sameerlite made their first contribution in #14460
• @holzman made their first contribution in #14459
• @sashank5644 made their first contribution in #14469
• @TomAlon made their first contribution in #14401
• @AlexsanderHamir made their first contribution in #14538

Full Changelog: v1.77.1.dev.2...v1.77.1-nightly

Docker Run LiteLLM Proxy

docker run \
-e STORE_MODEL_IN_DB=True \
-p 4000:4000 \
ghcr.io/berriai/litellm:main-v1.77.1-nightly

Don't want to maintain your internal proxy? get in touch 🎉

Hosted Proxy Alpha: https://calendly.com/d/4mp-gd3-k5k/litellm-1-1-onboarding-chat

Load Test LiteLLM Proxy Results