feat: OpenRouter provider and shared LiteLLM json_object allowlist#25
Merged
Merged
Conversation
Centralize OpenAI-style response_format providers for LangGraph/ADK parity, register openrouter in LLM config with OPENROUTER_API_KEY, and extend engine capability sets and tests. Co-authored-by: Cursor <cursoragent@cursor.com>
- Qwen 2.5: adds qwen-2.5-72b-instruct, qwen-2.5-7b-instruct-fp16, and qwen-2.5-coder-32b-instruct to Groq provider - Claude Haiku 4.5: adds claude-haiku-4-5 (→ claude-haiku-4-5-20251001) to the Claude provider - GPT-5: already present in OpenAI provider (confirmed) Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
New models added to existing providers: - OpenAI: o4-mini, gpt-4.5-preview - Groq: llama-4-scout-17b-16e-instruct, llama-4-maverick-17b-128e-instruct New provider: - DeepSeek: deepseek-chat (V3) and deepseek-reasoner (R1) via DeepSeek API Set DEEPSEEK_API_KEY to use. Note: GPT-5.5 does not exist yet — GPT-5 and o4-mini cover the latest OpenAI frontier. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Azure hosts OpenAI models behind a private endpoint. The llm_model value is the Azure deployment name (user-defined in Azure AI Studio), so the model list is open — any deployment name is accepted via LLMModel._missing_. Required env vars: AZURE_API_KEY, AZURE_API_BASE, AZURE_API_VERSION. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Collaborator
|
@Abhishek-Khanna24 Can we create a github issue to update the Readme.md and Documentation ? Also need to update the webapp. |
feat: add new LLM models and providers — Qwen 2.5, Claude Haiku 4.5, o4-mini, Llama 4, DeepSeek, Azure
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
llm_provider_name: openrouter) withOPENROUTER_API_KEY, LiteLLMopenrouter/model strings, and an open model catalog (same pattern as vLLM).json_object_response_formatas the single source of truth for which LiteLLM backends get OpenAI-styleresponse_format: json_object(includes OpenRouter), and wire LangGraph + ADK capability metadata accordingly.env.exampleand extend unit tests for OpenRouter and the shared allowlist.Test plan
pipenv run pytest tests/unit/agent_framework/configs/test_llm_provider_config.py -vOPENROUTER_API_KEYand smoke-test an agent withllm_provider_name: openrouterand a valid OpenRouter model slug on LangGraph and/or ADK.Made with Cursor