[Doc] Show how to get Responses prompt token IDs (#55912)

Signed-off-by: Francisco Javier Arceo <farceo@redhat.com>
Co-authored-by: OpenAI Codex <codex@openai.com>
This commit is contained in:
Francisco Javier Arceo
2026-09-16 02:48:18 +00:00
committed by GitHub
co-authored by OpenAI Codex
parent b0898a4937
commit 7d4bba5498
3 changed files with 44 additions and 3 deletions
+6 -1
View File
@@ -130,7 +130,12 @@ Scale-out APIs are disabled by default on `vllm serve`. Set `--enable-scale-out`
### Renderer APIs
For further details on renderer APIs, please refer to [this page](renderer.md).
Renderer APIs preprocess completion, chat, and Responses requests without running
inference. They handle tokenization, model-specific prompt formatting, and
multimodal preprocessing, returning prompt token IDs, sampling parameters, and
any processed multimodal inputs for generation.
See the [renderer guide](renderer.md) for setup instructions and examples.
- [Completions Render API](renderer.md) (`/v1/completions/render`)
- Render completion requests