Tau2-bench + vLLM本地部署调用

Tau2-bench vLLM本地部署调用背景1. vllm配置2. env配置3. 启动脚本4. 关闭下litellm模型计费设置5. 成功运行近期在运行 Tau2-bench 进行模型推理基准测试时发现其 GitHub 仓库的部署文档较为简略特别是缺少与 vLLM 框架结合进行本地模型部署的详细教程。因此本文旨在记录并分享在实际部署过程中遇到的典型问题与解决方案为有类似需求的开发者提供一份实用的踩坑指南。背景近期在运行 Tau2-bench 进行模型推理基准测试时发现其 GitHub 仓库的部署文档较为简略特别是缺少与 vLLM 框架结合进行本地模型部署的详细教程。因此本文旨在记录并分享在实际部署过程中遇到的典型问题与解决方案。1. vllm配置vllm启动时需要带入enable-auto-tool-choice和tool-call-parser如下所示vllm serve /root/models/Qwen3.5-2B \ --host 0.0.0.0 \ --port 8001 \ --served-model-name Qwen3.5-2B \ --enable-auto-tool-choice \ --tool-call-parser qwen3_coder2. env配置因为是调用本地API所以API_KEY都置空并关闭OpenRouter。ANTHROPIC_API_KEY OPENAI_API_KEY ELEVENLABS_API_KEY DEEPGRAM_API_KEY # Required for banking_knowledge qwen_embeddings* retrieval configs (via OpenRouter) #OPENROUTER_API_KEY # ── Voice Persona Overrides ──────────────────────────────────────────── # The default voice IDs are Sierra-internal and wont work for external users. # Create your own voices in ElevenLabs and set the IDs here. # See docs/voice-personas.md for a step-by-step guide. # # Control personas (American accents, used in control complexity): # TAU2_VOICE_ID_MATT_DELANEYyour_voice_id # TAU2_VOICE_ID_LISA_BRENNERyour_voice_id # # Regular personas (diverse accents, used in regular complexity): # TAU2_VOICE_ID_MILDRED_KAPLANyour_voice_id # TAU2_VOICE_ID_ARJUN_ROYyour_voice_id # TAU2_VOICE_ID_WEI_LINyour_voice_id # TAU2_VOICE_ID_MAMADOU_DIALLOyour_voice_id # TAU2_VOICE_ID_PRIYA_PATILyour_voice_id3. 启动脚本api_base改成本地endpoint--agent-llm和--user-llm需要改成openai/**本地启动模型ID**uv run tau2 run \ --agent-llm-args {api_key: , api_base: http://localhost:8001/v1} \ --user-llm-args {api_key: , api_base: http://localhost:8001/v1} \ --agent-llm openai/Qwen3.5-2B \ --user-llm openai/Qwen3.5-2B \ --domain telecom \ --num-trials 1 \ --num-tasks 54. 关闭下litellm模型计费设置完成步骤3后直接启动会报错如下所示但不影响运行因为tau2-bench需要对模型消耗token进行计费但是本地模型在litellm中并没有所以需要自己增加下相关的token计费信息或者直接在代码中关闭2026-06-05 02:24:09.848 | ERROR | tau2.utils.llm_utils:get_response_cost:129 - This model isnt mapped yet. modelQwen3.5-2B, custom_llm_provideropenai. Add it here - https://github.com/BerriAI/litellm/blob/main/model_prices_and_context_window.json. 2026-06-05 02:24:11.148 | ERROR | tau2.utils.llm_utils:get_response_cost:129 - This model isnt mapped yet. modelQwen3.5-2B, custom_llm_provideropenai. Add it here - https://github.com/BerriAI/litellm/blob/main/model_prices_and_context_window.json. 2026-06-05 02:24:11.231 | ERROR | tau2.utils.llm_utils:get_response_cost:129 - This model isnt mapped yet. modelQwen3.5-2B, custom_llm_provideropenai. Add it here - https://github.com/BerriAI/litellm/blob/main/model_prices_and_context_window.json. 2026-06-05 02:24:11.577 | ERROR | tau2.utils.llm_utils:get_response_cost:129 - This model isnt mapped yet. modelQwen3.5-2B, custom_llm_provideropenai. Add it here - https://github.com/BerriAI/litellm/blob/main/model_prices_and_context_window.json.解决方法进入对应tau2-bench/src/tau2/utils/llm_utils.py修改注释下129行119 def get_response_cost(response: ModelResponse) - float: 120 121 Get the cost of the response from the litellm completion. 122 123 response.model _parse_ft_model_name( 124 response.model 125 ) # FIXME: Check Litellm, passing the model to completion_cost doesnt work. 126 try: 127 cost completion_cost(completion_responseresponse) 128 except Exception as e: 129 #logger.error(e) 130 return 0.0 131 return cost5. 成功运行