Cesar Garcia

Hi! This would be a great feature to have! In my case, I’m currently not receiving the confirmation email, so having this implemented would really help. Is there any update?...

> [@srameshr](https://github.com/srameshr) sure. let's keep it subtle. Here is a reference design > >  > > [@nithindavid](https://github.com/nithindavid) any comments on the design? Hi!...

Just closed this PR. Tried it multiple models: qwen3:0.6b, qwen3:8b/phi3:3.8b (decent quality but 10-30s per call). Larger models like llama3.1:70b could work but require 16GB+ RAM, high-end CPUs/GPUs, and still...

Thanks for checking, @mamoodi The PR #11403 removed the broken PNG reference , so the README points to the Mermaid diagrams in the docs site—but those diagrams reflect the V0...

Hi @mamoodi Great. Diagramas of the v1 architecture are made before the software or while or when v1 the versión is finished?

> Looks like anthropic thinking is incompatible with OpenAI compatiple apis Yeah. This is a known limitation due to fundamental architectural differences between OpenAI and Anthropic APIs. **OpenAI and Anthropic...

Hi @szafranek! This was fixed in PR #16745 and is already merged to `main`. The fix will be included in the next release. You can: Install from main: `pip install...

> Thanks [@Chesars](https://github.com/Chesars)! I missed that PR when I filed the issue since it didn't mention streaming, but I'm happy the problem been already fixed! Awesome @szafranek 🙌

> [@Chesars](https://github.com/Chesars) can you please look into fixing this? Cool, PR https://github.com/BerriAI/litellm/pull/18092 fixes

> @Chesars Could you please update the PR by resolving the conflicts? Hi @iamsivin. There I updated the branch.