refactor(vllm): own nemotron thinking payloads

This commit is contained in:
Peter Steinberger
2026-04-27 12:13:37 +01:00
parent 22bb53ac9a
commit da822a56d8
11 changed files with 244 additions and 160 deletions

View File

@@ -153,7 +153,7 @@ Use explicit config when:
<Accordion title="Nemotron 3 thinking controls">
vLLM/Nemotron 3 can use chat-template kwargs to control whether reasoning is
returned as hidden reasoning or visible answer text. When an OpenClaw session
uses `vllm/nemotron-3-*` with thinking off, OpenClaw sends:
uses `vllm/nemotron-3-*` with thinking off, the bundled vLLM plugin sends:
```json
{