EngineX-Iluvatar/enginex-vllm-bi100-qwen36
16
0
Fork 0
Code Issues Pull Requests Actions Projects Releases Wiki Activity

Add qwen3_coder tool calling parser

Browse Source
This commit is contained in:
Lu Xinlong
2026-06-10 14:38:54 +08:00
parent d972854fb7
commit 4ab36b51d5
3 changed files with 595 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

7
qwen3_6_scripts/patch_ops.sh
View File

@@ -51,3 +51,10 @@ cp ./sequence.py /usr/local/corex/lib/python3/dist-packages/vllm/sequence.py
# works correctly during profiling runs with chunked-prefill-style batches.
# also bypasses auto chunked prefill on
python3 ./patch_xformers_sdpa_seq.py
# --- tool parser: Qwen3 XML tool call format ---------------------------------
# Registers "qwen3_coder" parser for Qwen3.6 XML-style tool calls:
# <tool_call><function=name><parameter=key>\nvalue\n</parameter></function></tool_call>
# Use at server start: --tool-call-parser qwen3_coder --enable-auto-tool-choice
cp ./qwen3coder_tool_parser.py /usr/local/corex/lib/python3/dist-packages/vllm/entrypoints/openai/tool_parsers/
python3 ./patch_vllm_tool_parser.py