This release adds 3 notable features for engineering teams evaluating rollout.
Published 2mo
LLM Frameworks
✓ No known CVEs patched
✓ No known CVEs patched in this version
Topics
ai
ai-gateway
ai-governance
ai-proxy
ai-safety
ai-security
+14 more
ai-tools
anthropic
deepseek
litellm-alternative
llm
llm-gateway
observability
ollama
openai
openai-compatible-api
openai-proxy
opencode-zen
proxy
zhipu-ai
Summary
AI summaryImplement structured error handling with an error_kind field.
Full changelog
Changelog
- 22794209d4c1029067fb5e445bfd83b3a8a4e4a2 Add code review follow-ups: Bouncer test and comment fix
- a374c75c7711d3ba3f3d98de04492fee4e5a975d Fix interceptor CanHandle logic when MaxContext is unknown
- aff61ab06ce04b878820cec50b1100fe59668cc5 feat: 002 implement local engine lifecycle
- 6380e42541d5d1f27e90e72654c6d6793b9e8e2c feat: add bidirectional Anthropic Messages API support
- 8f0dc7792dfe78828bc7ac041c9118b8882a9a98 feat: add context-limit aware retry with summarization fallback
- e343fb9459348f609714b2a2ea2a6d8b671bd286 feat: add metrics and logging instrumentation for post-v0.3.0 features
- 277277933a51efa8d4c0154b9ffdc51c011437bb feat: enrich Ollama model discovery with /api/show metadata
- c6b4c17ff0e0b8185cca65329d48aaea89a88233 feat: implement pluggable interceptor chain for content preprocessing
- 81df23e61beebedfffee83721406d064ccc74a14 feat: implement structured error handling with error_kind field
- 1b1fdae24c63f240e353a4c0019f03372f14243b feat: strip unsupported tool_choice field from Ollama requests
- 21f0016a440916da29df0d8241adb851a5a05d92 fix: merge built-in provider defaults into partial user configs
Weekly OSS security release digest.
The CVE patches and breaking changes that affected production tools this week. One email, every Sunday.
No spam, unsubscribe anytime.
Share this release
About Nenya
All releases →Related context
Related tools
Beta — feedback welcome: [email protected]