> 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.

# Recipes

## Docs

- [Recipes](https://docs.nvidia.com/dynamo/dev/recipes/browse.md): Pick a model, choose the hardware/runtime target, and deploy the matching Dynamo recipe.
- [Motif-3 NVFP4](https://docs.nvidia.com/dynamo/dev/recipes/motif-3.md): Serve Motif-3 NVFP4 with Dynamo and the Motif vLLM runtime on NVIDIA B200 GPUs.
- [GLM-5.3-Flash](https://docs.nvidia.com/dynamo/dev/recipes/glm-5-3-flash.md): Serve GLM-5.3-Flash with Dynamo and vLLM on GB200 or H200, aggregated or with prefill/decode disaggregation.
- [GLM-5.3/5.2](https://docs.nvidia.com/dynamo/dev/recipes/glm-5-2.md): Serve GLM-5.3/5.2 with Dynamo and SGLang on B200 or H200 for long-context agentic workloads.
- [GLM-5 NVFP4](https://docs.nvidia.com/dynamo/dev/recipes/glm-5-nvfp4.md): Serve GLM-5 NVFP4 with Dynamo and SGLang, disaggregated across five GB200 nodes.
- [Kimi-K3](https://docs.nvidia.com/dynamo/dev/recipes/kimi-k3.md): Serve Kimi-K3 with Dynamo on H200, GB200, or GB300.
- [Kimi-K2.6](https://docs.nvidia.com/dynamo/dev/recipes/kimi-k2-6.md): Serve moonshotai/Kimi-K2.6 with Dynamo and vLLM, tuned per GPU and workload.
- [MiniMax M3 NVFP4](https://docs.nvidia.com/dynamo/dev/recipes/minimax-m3.md): Serve MiniMax M3 NVFP4 with Dynamo and vLLM on GB200, aggregated or with prefill/decode disaggregation.
- [DeepSeek-V4.1-Flash](https://docs.nvidia.com/dynamo/dev/recipes/deepseek-v4-1-flash.md): Serve DeepSeek-V4.1-Flash with Dynamo on B200, GB200, and H200 using aggregated or disaggregated workers.
- [DeepSeek-V4-Pro](https://docs.nvidia.com/dynamo/dev/recipes/deepseek-v4-pro.md): Serve DeepSeek-V4-Pro with Dynamo and vLLM — aggregated or disaggregated, on B200 or H200, tuned for the agentic workload.
- [DeepSeek-V4-Pro-0813](https://docs.nvidia.com/dynamo/dev/recipes/deepseek-v4-pro-0813.md): Serve DeepSeek-V4-Pro-0813 with Dynamo and vLLM on GB200 or H200, aggregated or disaggregated, with 1M context.
- [DeepSeek-V4-Flash](https://docs.nvidia.com/dynamo/dev/recipes/deepseek-v4-flash.md): Serve DeepSeek-V4-Flash with Dynamo and vLLM — aggregated or disaggregated, on B200 or H200, tuned for the agentic workload.
- [Qwen3.8-Flash-Next](https://docs.nvidia.com/dynamo/dev/recipes/qwen-3-8-flash-next.md): Serve Qwen3.8-Flash-Next with Dynamo and vLLM on B200, aggregated or disaggregated.
- [Qwen3.8-2.4T-A95B](https://docs.nvidia.com/dynamo/dev/recipes/qwen-3-8-2-4t-a95b-fp8.md): Serve Qwen3.8-2.4T-A95B with Dynamo and vLLM or SGLang on GB300 or GB200, aggregated or disaggregated.
- [Qwen3.6-35B-A3B](https://docs.nvidia.com/dynamo/dev/recipes/qwen3-6-35b-a3b.md): Serve Qwen3.6-35B-A3B with Dynamo on B200, GB200, or H200.
- [Qwen3.5-122B-A10B](https://docs.nvidia.com/dynamo/dev/recipes/qwen3-5-122b.md): Serve Qwen3.5-122B-A10B with Dynamo and vLLM on B200 or H200 for long-context agentic workloads.
- [Qwen3-235B-A22B FP8](https://docs.nvidia.com/dynamo/dev/recipes/qwen3-235b-a22b-fp8.md): Serve Qwen/Qwen3-235B-A22B-FP8 with Dynamo and TensorRT-LLM on 16 Hopper or Blackwell GPUs.
- [Qwen3-32B](https://docs.nvidia.com/dynamo/dev/recipes/qwen3-32b.md): Serve Qwen3-32B with Dynamo and vLLM — disaggregated serving with KV-aware routing for multi-turn conversational traffic.
- [Qwen3-32B FP8](https://docs.nvidia.com/dynamo/dev/recipes/qwen3-32b-fp8.md): Serve Qwen/Qwen3-32B-FP8 with Dynamo on TensorRT-LLM or vLLM, aggregated or disaggregated.
- [Nemotron 3.5 Lightning](https://docs.nvidia.com/dynamo/dev/recipes/nemotron-3-5-lightning.md): Serve NVIDIA-Nemotron-3.5-Lightning-30B-A3B in NVFP4 or BF16 with Dynamo, vLLM, and experimental TensorRT-LLM targets.
- [Nemotron-3-Ultra](https://docs.nvidia.com/dynamo/dev/recipes/nemotron-3-ultra.md): Serve Nemotron-3-Ultra with optimized Dynamo 1.4.0 and vLLM profiles on B200, GB200, and H200.
- [Nemotron-3-Super](https://docs.nvidia.com/dynamo/dev/recipes/nemotron-3-super.md): Serve NVIDIA-Nemotron-3-Super-120B-A12B with Dynamo and vLLM, tuned per GPU and workload.
- [Gemma-4-31B](https://docs.nvidia.com/dynamo/dev/recipes/gemma-4-31b.md): Serve Gemma-4-31B with Dynamo and TensorRT-LLM on B200, GB200, or H200 for long-context agentic workloads.
- [Inkling NVFP4](https://docs.nvidia.com/dynamo/dev/recipes/inkling.md): Serve Thinking Machines' Inkling on GB300 with vLLM for long-context agentic workloads, or on B200 with SGLang for multimodal input.
- [GPT-OSS-120B](https://docs.nvidia.com/dynamo/dev/recipes/gpt-oss-120b.md): Serve openai/gpt-oss-120b with Dynamo on vLLM (B200/H200) or TensorRT-LLM (GB200), aggregated or disaggregated.
- [K-EXAONE 2.0](https://docs.nvidia.com/dynamo/dev/recipes/k-exaone-2.md): Serve LG AI Research's K-EXAONE 2.0 750B-A37B in NVFP4 with Dynamo and vLLM on B200.
- [A.X-K2](https://docs.nvidia.com/dynamo/dev/recipes/ax-k2.md): Serve A.X-K2 with Dynamo and vLLM on B200 or H200, aggregated or disaggregated.
- [Solar Open2 250B](https://docs.nvidia.com/dynamo/dev/recipes/solar-open2-250b.md): Serve Solar-Open2-250B-NVFP4 with Dynamo on B200, aggregated or disaggregated.
- [SGLang Deployment Templates](https://docs.nvidia.com/dynamo/dev/recipes/kubernetes-templates/dgd/sg-lang.md): Ready-to-apply DynamoGraphDeployment manifests for serving SGLang with Dynamo on Kubernetes.
- [TensorRT-LLM Deployment Templates](https://docs.nvidia.com/dynamo/dev/recipes/kubernetes-templates/dgd/tensor-rt-llm.md): Ready-to-apply DynamoGraphDeployment manifests for serving TensorRT-LLM with Dynamo on Kubernetes.
- [vLLM Deployment Templates](https://docs.nvidia.com/dynamo/dev/recipes/kubernetes-templates/dgd/v-llm.md): Ready-to-apply DynamoGraphDeployment manifests for serving vLLM with Dynamo on Kubernetes.
- [DGDR Templates](https://docs.nvidia.com/dynamo/dev/recipes/kubernetes-templates/dgdr.md): Ready-to-apply DynamoGraphDeploymentRequest manifests for profiling and generating DynamoGraphDeployments.
- [SGLang Local Deployment Examples](https://docs.nvidia.com/dynamo/dev/recipes/cli-templates/sg-lang.md): Copyable commands for running the SGLang launch scripts in the Dynamo repository.
- [TensorRT-LLM Local Deployment Examples](https://docs.nvidia.com/dynamo/dev/recipes/cli-templates/tensor-rt-llm.md): Copyable commands for running the TensorRT-LLM launch scripts in the Dynamo repository.
- [vLLM Local Deployment Examples](https://docs.nvidia.com/dynamo/dev/recipes/cli-templates/v-llm.md): Copyable commands for running the vLLM launch scripts in the Dynamo repository.
- [Feature Benchmarks](https://docs.nvidia.com/dynamo/dev/recipes/benchmarks.md): Start with the feature or performance claim you want to inspect.