@tduncan / Nvidia-Optimization-Stack-VLLM Mirror · read-only / commits / 70ca5484f5

Fork 0

A high-throughput and memory-efficient inference and serving engine for LLMs

Mirrored from https://github.com/tduncan-lum/Nvidia-Optimization-Stack---VLLM.git

Synced Nov 15, 02:15 UTC

[Doc] Update notes (#20668)

Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>

Cyrus Leung committed Jul 9, 2025 at 18:46 UTC 70ca5484f5b5b11c3a3d0811b3c12a3e795f5655

3 files changed +14 -14

docs/deployment/integrations/production-stack.md

+6 -3

@@ -41,7 +41,8 @@ vllm-deployment-router-859d8fb668-2x2b7 1/1 Running 0 2m38

41 vllm-opt125m-deployment-vllm-84dfc9bd7-vb9bs 1/1 Running 0 2m38s

42 ```

44 -**NOTE**: It may take some time for the containers to download the Docker images and LLM weights.

44 +!!! note

45 + It may take some time for the containers to download the Docker images and LLM weights.

47 ### Send a Query to the Stack

@@ -149,6 +150,8 @@ In this YAML configuration:

150 * **`requestGPU`**: Specifies the number of GPUs required.

151 * **`pvcStorage`**: Allocates persistent storage for the model.

152

152 -**NOTE:** If you intend to set up two pods, please refer to this [YAML file](https://github.com/vllm-project/production-stack/blob/main/tutorials/assets/values-01-2pods-minimal-example.yaml).

153 +!!! note

154 + If you intend to set up two pods, please refer to this [YAML file](https://github.com/vllm-project/production-stack/blob/main/tutorials/assets/values-01-2pods-minimal-example.yaml).

155

154 -**NOTE:** vLLM production stack offers many more features (*e.g.* CPU offloading and a wide range of routing algorithms). Please check out these [examples and tutorials](https://github.com/vllm-project/production-stack/tree/main/tutorials) and our [repo](https://github.com/vllm-project/production-stack) for more details!

156 +!!! tip

157 + vLLM production stack offers many more features (*e.g.* CPU offloading and a wide range of routing algorithms). Please check out these [examples and tutorials](https://github.com/vllm-project/production-stack/tree/main/tutorials) and our [repo](https://github.com/vllm-project/production-stack) for more details!

docs/features/tool_calling.md

+6 -9

@@ -299,20 +299,17 @@ Limitations:

299

300 Example supported models:

301

302 -* `meta-llama/Llama-3.2-1B-Instruct`\* (use with <gh-file:examples/tool_chat_template_llama3.2_pythonic.jinja>)

303 -* `meta-llama/Llama-3.2-3B-Instruct`\* (use with <gh-file:examples/tool_chat_template_llama3.2_pythonic.jinja>)

302 +* `meta-llama/Llama-3.2-1B-Instruct` ⚠️ (use with <gh-file:examples/tool_chat_template_llama3.2_pythonic.jinja>)

303 +* `meta-llama/Llama-3.2-3B-Instruct` ⚠️ (use with <gh-file:examples/tool_chat_template_llama3.2_pythonic.jinja>)

304 * `Team-ACE/ToolACE-8B` (use with <gh-file:examples/tool_chat_template_toolace.jinja>)

305 * `fixie-ai/ultravox-v0_4-ToolACE-8B` (use with <gh-file:examples/tool_chat_template_toolace.jinja>)

306 -* `meta-llama/Llama-4-Scout-17B-16E-Instruct`\* (use with <gh-file:examples/tool_chat_template_llama4_pythonic.jinja>)

307 -* `meta-llama/Llama-4-Maverick-17B-128E-Instruct`\* (use with <gh-file:examples/tool_chat_template_llama4_pythonic.jinja>)

306 +* `meta-llama/Llama-4-Scout-17B-16E-Instruct` ⚠️ (use with <gh-file:examples/tool_chat_template_llama4_pythonic.jinja>)

307 +* `meta-llama/Llama-4-Maverick-17B-128E-Instruct` ⚠️ (use with <gh-file:examples/tool_chat_template_llama4_pythonic.jinja>)

308

309 Flags: `--tool-call-parser pythonic --chat-template {see_above}`

310

311 ----

312 -**WARNING**

313 -Llama's smaller models frequently fail to emit tool calls in the correct format. Your mileage may vary.

314 -

315 ----

311 +!!! warning

312 + Llama's smaller models frequently fail to emit tool calls in the correct format. Your mileage may vary.

313

314 ## How to write a tool parser plugin

315

docs/models/supported_models.md

+2 -2

@@ -573,7 +573,7 @@ Specified using `--task generate`.

573 | `GLM4VForCausalLM`^ | GLM-4V | T + I | `THUDM/glm-4v-9b`, `THUDM/cogagent-9b-20241220`, etc. | ✅︎ | ✅︎ | ✅︎ |

574 | `Glm4vForConditionalGeneration` | GLM-4.1V-Thinking | T + IE+ + VE+ | `THUDM/GLM-4.1V-9B-Thinkg`, etc. | ✅︎ | ✅︎ | ✅︎ |

575 | `GraniteSpeechForConditionalGeneration` | Granite Speech | T + A | `ibm-granite/granite-speech-3.3-8b` | ✅︎ | ✅︎ | ✅︎ |

576 -| `H2OVLChatModel` | H2OVL | T + IE+ | `h2oai/h2ovl-mississippi-800m`, `h2oai/h2ovl-mississippi-2b`, etc. | | ✅︎ | ✅︎\* |

576 +| `H2OVLChatModel` | H2OVL | T + IE+ | `h2oai/h2ovl-mississippi-800m`, `h2oai/h2ovl-mississippi-2b`, etc. | | ✅︎ | ✅︎ |

577 | `Idefics3ForConditionalGeneration` | Idefics3 | T + I | `HuggingFaceM4/Idefics3-8B-Llama3`, etc. | ✅︎ | | ✅︎ |

578 | `InternVLChatModel` | InternVL 3.0, InternVideo 2.5, InternVL 2.5, Mono-InternVL, InternVL 2.0 | T + IE+ + (VE+) | `OpenGVLab/InternVL3-9B`, `OpenGVLab/InternVideo2_5_Chat_8B`, `OpenGVLab/InternVL2_5-4B`, `OpenGVLab/Mono-InternVL-2B`, `OpenGVLab/InternVL2-4B`, etc. | ✅︎ | ✅︎ | ✅︎ |

@@ -599,7 +599,7 @@ Specified using `--task generate`.

599 | `Qwen2AudioForConditionalGeneration` | Qwen2-Audio | T + A+ | `Qwen/Qwen2-Audio-7B-Instruct` | | ✅︎ | ✅︎ |

600 | `Qwen2VLForConditionalGeneration` | QVQ, Qwen2-VL | T + IE+ + VE+ | `Qwen/QVQ-72B-Preview`, `Qwen/Qwen2-VL-7B-Instruct`, `Qwen/Qwen2-VL-72B-Instruct`, etc. | ✅︎ | ✅︎ | ✅︎ |

601 | `Qwen2_5_VLForConditionalGeneration` | Qwen2.5-VL | T + IE+ + VE+ | `Qwen/Qwen2.5-VL-3B-Instruct`, `Qwen/Qwen2.5-VL-72B-Instruct`, etc. | ✅︎ | ✅︎ | ✅︎ |

602 -| `Qwen2_5OmniThinkerForConditionalGeneration` | Qwen2.5-Omni | T + IE+ + VE+ + A+ | `Qwen/Qwen2.5-Omni-7B` | | ✅︎ | ✅︎\* |

602 +| `Qwen2_5OmniThinkerForConditionalGeneration` | Qwen2.5-Omni | T + IE+ + VE+ + A+ | `Qwen/Qwen2.5-Omni-7B` | | ✅︎ | ✅︎ |

603 | `SkyworkR1VChatModel` | Skywork-R1V-38B | T + I | `Skywork/Skywork-R1V-38B` | | ✅︎ | ✅︎ |

604 | `SmolVLMForConditionalGeneration` | SmolVLM2 | T + I | `SmolVLM2-2.2B-Instruct` | ✅︎ | | ✅︎ |

605 | `TarsierForConditionalGeneration` | Tarsier | T + IE+ | `omni-search/Tarsier-7b`, `omni-search/Tarsier-34b` | | ✅︎ | ✅︎ |