added response_template
This PR adds support for the response_template in tokenizer_config.json. A response template tells the tokenizer how to parse the model's raw output into a structured assistant message (reasoning_content, content, tool_calls), so tokenizer.parse_response() and chat pipelines return parsed messages instead of raw text. See https://huggingface.co/docs/transformers/en/chat_response_parsing
Template
This is the existing qwen3 template, copied unchanged from onnx-community/Qwen3-0.6B-ONNX. This repo is an ONNX export of HuggingFaceTB/SmolLM3-3B with the same chat template and output format: <think>…</think> reasoning and JSON <tool_call> blocks. The original model has no response_template of its own.
How it was tested
We ran this checkpoint (q4f16) with Transformers.js, using its own chat template, on a single tool call, two parallel tool calls, a plain answer, a plain question with a tool available, and a run with thinking disabled. Every well-formed output parsed correctly, both in one shot and when streamed. Upstream Python transformers gives the same results on the same outputs.
The template describes the checkpoint's output format. It does not mean that every response contains every region, and it does not guarantee tool-use quality.