AGENTS.md

1### Model Factory (`packages/harness/deerflow/models/factory.py`)
2 
3- `create_chat_model(name, thinking_enabled)` instantiates LLM from config via reflection
4- Supports `thinking_enabled` flag with per-model `when_thinking_enabled` overrides
5- Supports vLLM-style thinking toggles via `when_thinking_enabled.extra_body.chat_template_kwargs.enable_thinking` for Qwen reasoning models, while normalizing legacy `thinking` configs for backward compatibility
6- Supports `supports_vision` flag for image understanding models
7- Config values starting with `$` resolved as environment variables
8- Missing provider modules surface actionable install hints from reflection resolvers (for example `uv add langchain-google-genai`)
9 
10### vLLM Provider (`packages/harness/deerflow/models/vllm_provider.py`)
11 
12- `VllmChatModel` subclasses `langchain_openai:ChatOpenAI` for vLLM 0.19.0 OpenAI-compatible endpoints
13- Preserves vLLM's non-standard assistant `reasoning` field on full responses, streaming deltas, and follow-up tool-call turns
14- Designed for configs that enable thinking through `extra_body.chat_template_kwargs.enable_thinking` on vLLM 0.19.0 Qwen reasoning models, while accepting the older `thinking` alias
15- `cumulative_stream_usage` is an opt-in model setting (default `false`) for endpoints that repeat cumulative token totals on each streaming chunk. The provider converts snapshots to deltas only when a stable completion id is present, isolates interleaved streams by id, and leaves the original usage untouched otherwise. Per-model tracking is lock-protected and cleared on the trailing empty-`choices` frame whether or not that frame carries usage. A soft cap of 1024 ids evicts only entries idle for at least one hour; active streams may temporarily exceed the cap so eviction cannot corrupt their deltas. Regression coverage lives in `tests/test_vllm_provider.py`.
16