> For clean Markdown content of this page, append .md to this URL. For the complete documentation index, see https://docs.nvidia.com/dynamo/llms.txt. For full content including API reference and SDK examples, see https://docs.nvidia.com/dynamo/llms-full.txt. # Tool Call Parsing (Dynamo) Tool calling lets Dynamo connect models to external tools and services by returning function arguments for your application to execute. Requests use the `tool_choice` and `tools` parameters. This page covers Dynamo-native parsing; if your model is not listed, use [engine fallback](/dynamo/dev/parsing/chat-processors), which also explains valid combinations of `--dyn-tool-call-parser`, `--dyn-chat-processor`, and `--dyn-reasoning-parser`. ## Configure Tool Calling #### Launch Dynamo Backend Select a `--dyn-tool-call-parser` that matches your model based on the list below. The backend can be SGLang, TensorRT-LLM, vLLM, or another installed backend. To inspect the available worker flags, run: ```bash python -m dynamo. --help ``` The following example starts an SGLang worker for Qwen3.5: ```bash python -m dynamo.sglang \ --model Qwen/Qwen3.5-4B \ --dyn-tool-call-parser qwen3_coder \ --reasoning-parser qwen3 \ --dyn-reasoning-parser qwen3 ``` If the model supports tool calling but its default chat template does not, add `--custom-jinja-template /path/to/template.jinja` to the worker command. The template must be readable inside the container. If the model emits reasoning content that should be separated from normal output, see [Reasoning Parsing (Dynamo)](/dynamo/dev/parsing/reasoning-parsing) for supported `--dyn-reasoning-parser` values. #### Start the Frontend Start the Dynamo frontend in another terminal: ```bash python -m dynamo.frontend ``` #### Wait for Readiness Wait for the frontend health endpoint to return a successful response: ```bash curl --fail http://localhost:8000/health ``` #### Send a Tool-Calling Request Send an OpenAI-compatible chat completion request with the available tools: ```bash curl -s http://localhost:8000/v1/chat/completions \ -H 'Content-Type: application/json' \ -d '{ "model": "Qwen/Qwen3.5-4B", "messages": [ {"role": "user", "content": "What is the weather in San Francisco and New York?"} ], "tools": [{ "type": "function", "function": { "name": "get_weather", "description": "Get the current weather for a location.", "parameters": { "type": "object", "properties": {"location": {"type": "string"}}, "required": ["location"] } } }], "tool_choice": "auto" }' ``` Dynamo parses the model output and returns OpenAI-compatible `tool_calls`. #### View an example response ```json { "id": "chatcmpl-b415caad-9be0-4d9e-ac6d-9d23bfe57703", "choices": [ { "index": 0, "message": { "role": "assistant", "content": null, "reasoning_content": "The user is asking about the weather in two cities: San Francisco and New York. I need to call the get_weather function for each city. I'll make two separate function calls to get the weather information for both locations.\n", "tool_calls": [ { "id": "call-56223a95-3d14-4433-a94e-011f106c0e40", "type": "function", "function": { "name": "get_weather", "arguments": "{\"location\":\"San Francisco\"}" } }, { "id": "call-d5b5772b-6b0c-4120-ad01-623278a937fe", "type": "function", "function": { "name": "get_weather", "arguments": "{\"location\":\"New York\"}" } } ] }, "finish_reason": "tool_calls", "logprobs": null } ], "created": 1778653281, "model": "Qwen/Qwen3.5-4B", "object": "chat.completion" } ``` If a tool call is parsed incorrectly, add `"logprobs": true` to one reproduction request and share the response. See [Troubleshooting Tool Calls](/dynamo/dev/parsing/troubleshooting-tool-calls) for what to capture when reporting an issue. ## Supported Tool Call Parsers Choose a model family, then expand the matching model option to see its parser name and configuration details. The **Upstream name** column shows where the vLLM or SGLang parser name differs from Dynamo’s — relevant when using `--dyn-chat-processor vllm` or `sglang` (see [Parser Engine Fallback](/dynamo/dev/parsing/chat-processors)). A blank upstream column means the same name works everywhere. `Dynamo-only` means no upstream parser exists for this format. #### Kimi #### Kimi K2 Instruct/Thinking and Kimi K2.5 **Parser:** `kimi_k2` **Upstream name:** Same name across Dynamo, vLLM, and SGLang **Notes:** Pair with `--dyn-reasoning-parser kimi` or `kimi_k25`. For Kimi K2.5 thinking models, use `kimi_k25` so `` blocks and tool calls are parsed from the same response. See [Reasoning Parsing (Dynamo)](/dynamo/dev/parsing/reasoning-parsing). #### MiniMax #### MiniMax M2 and M2.1 **Parser:** `minimax_m2` **Upstream name:** vLLM: `minimax_m2` **Notes:** XML `` #### MiniMax M3 **Parser:** `minimax_m3` **Upstream name:** vLLM: `minimax_m3` **Notes:** MiniMax namespace-token XML #### DeepSeek #### DeepSeek V4 Pro / Flash **Parser:** `deepseek_v4` **Upstream name:** vLLM: `deepseek_v4`; SGLang: `deepseekv4` **Notes:** DSML tags (`<｜DSML｜tool_calls>...`). Aliases: `deepseek-v4`, `deepseekv4` #### DeepSeek V3 and DeepSeek R1-0528+ **Parser:** `deepseek_v3` **Upstream name:** SGLang: `deepseekv3` **Notes:** Special Unicode markers #### DeepSeek V3.1 **Parser:** `deepseek_v3_1` **Upstream name:** Dynamo-only **Notes:** JSON separators #### DeepSeek V3.2+ **Parser:** `deepseek_v3_2` **Upstream name:** Dynamo-only **Notes:** DSML tags (`<｜DSML｜function_calls>...`) #### Qwen, QwQ, and NousHermes #### Qwen3.5 and Qwen3-Coder **Parser:** `qwen3_coder` **Upstream name:** Same name across Dynamo, vLLM, and SGLang **Notes:** XML `` #### Qwen2.5-\*, QwQ-32B, Qwen3-Instruct, Qwen3-Think, and NousHermes-2/3 **Parser:** `hermes` **Upstream name:** vLLM: `qwen2_5`; SGLang: `qwen25` (for Qwen models) **Notes:** `` JSON #### GLM #### GLM-4.5 and GLM-4.7 **Parser:** `glm47` **Upstream name:** Dynamo-only **Notes:** XML `/` #### Nemotron #### Nemotron-Super / -Ultra / -Deci and Llama-Nemotron-Ultra / -Super **Parser:** `nemotron_deci` **Upstream name:** Dynamo-only **Notes:** `` JSON #### Nemotron-Nano **Parser:** `nemotron_nano` **Upstream name:** Dynamo-only **Notes:** Alias for `qwen3_coder` #### Gemma #### Google Gemma 4 thinking models **Parser:** `gemma4` **Upstream name:** vLLM: `gemma4` **Notes:** Custom non-JSON grammar with `<\|"\|>` string delimiters and `<\|tool_call>...` markers. Aliases: `gemma-4`. Pair with `--dyn-reasoning-parser gemma4` and `--custom-jinja-template examples/chat_templates/gemma4_tool.jinja`. #### gpt-oss #### gpt-oss-20b and gpt-oss-120b **Parser:** `harmony` **Upstream name:** Dynamo-only **Notes:** Harmony channel format #### Phi #### Phi-4, Phi-4-mini, and Phi-4-mini-reasoning **Parser:** `phi4` **Upstream name:** vLLM: `phi4_mini_json` **Notes:** `functools[...]` JSON #### Llama #### Llama 4 Scout / Maverick **Parser:** `pythonic` **Upstream name:** Same name across Dynamo, vLLM, and SGLang **Notes:** Python-list tool syntax #### Llama 3 / 3.1 / 3.2 / 3.3 Instruct **Parser:** `llama3_json` **Upstream name:** Same name across Dynamo, vLLM, and SGLang **Notes:** `<\|python_tag\|>` tool syntax #### Mistral #### Mistral, Mixtral, Mistral-Nemo, and Magistral **Parser:** `mistral` **Upstream name:** Same name across Dynamo, vLLM, and SGLang **Notes:** `[TOOL_CALLS]...[/TOOL_CALLS]` #### Jamba #### Jamba 1.5 / 1.6 / 1.7 **Parser:** `jamba` **Upstream name:** Dynamo-only **Notes:** `` JSON #### Fallback #### Default parser **Parser:** `default` **Upstream name:** Dynamo-only **Notes:** Empty JSON config (no start/end tokens). Prefer a model-specific parser for production use. #### Optional: structural tags You can enable **xgrammar structural tags** so guided decoding matches the parser's tool-call format at token granularity. See [Structural Tag (Guided Decoding for Tool Calls)](/dynamo/dev/parsing/structural-tag).