> For the complete documentation index, see [llms.txt](https://docs.modular.com/llms.txt).
> Markdown versions of all pages are available by appending .md to any URL (e.g. /max/get-started.md).

# Supported models

The table below lists all the model architectures currently supported by MAX.

Each model architecture represents a family of different models, as defined by
Hugging Face Transformers. The example model names are Hugging Face repository
IDs, such as `google/gemma-3-27b-it` for the `Gemma3ForCausalLM` architecture,
but you can use any model from Hugging Face that's based on an architecture
below.

To deploy any of these models with MAX, pass the model name to the [`max
serve`](https://docs.modular.com/max/cli/serve.md) or [`docker run`](https://docs.modular.com/max/container.md) command. Try it now
by following the [MAX quickstart guide](https://docs.modular.com/max/get-started.md). Or if you want to
serve a custom model, see the tutorial to [serve custom model
architectures](https://docs.modular.com/max/develop/serve-custom-model-architectures.md).

You can also see the
[model source code in GitHub](https://github.com/modular/modular/tree/main/max/python/max/pipelines/architectures).

<table className="models-table">
  <thead>
    <tr>
      <th>Architecture</th>
      <th>Example models (repo IDs)</th>
      <th>Modality</th>
      <th>Encodings</th>
      <th>Multi-GPU</th>
    </tr>
  </thead>
  <tbody>
    <tr>
      <td className='arch'><a href="https://github.com/modular/modular/tree/main/max/python/max/pipelines/architectures/bert"><code>BertModel</code></a></td>
      <td className='models'>
        <a href="https://huggingface.co/sentence-transformers/all-MiniLM-L6-v2">sentence-transformers/all-MiniLM-L6-v2</a>,<br/>
        <a href="https://huggingface.co/sentence-transformers/all-MiniLM-L12-v2">sentence-transformers/all-MiniLM-L12-v2</a>
      </td>
      <td className='modality'>text-to-embeddings</td>
      <td className='encodings'>bfloat16, float32</td>
      <td className='gpus'>No</td>
    </tr>
    <tr>
      <td className='arch'><a href="https://github.com/modular/modular/tree/main/max/python/max/pipelines/architectures/deepseekV2"><code>DeepseekV2ForCausalLM</code></a></td>
      <td className='models'>
        <a href="https://huggingface.co/deepseek-ai/DeepSeek-V2-Lite-Chat">deepseek-ai/DeepSeek-V2-Lite-Chat</a>
      </td>
      <td className='modality'>text-to-text</td>
      <td className='encodings'>bfloat16</td>
      <td className='gpus'>Yes</td>
    </tr>
    <tr>
      <td className='arch'><a href="https://github.com/modular/modular/tree/main/max/python/max/pipelines/architectures/deepseekV3_2"><code>DeepseekV32ForCausalLM</code></a></td>
      <td className='models'>
        <a href="https://huggingface.co/deepseek-ai/DeepSeek-V3.2">deepseek-ai/DeepSeek-V3.2</a>,<br/>
        <a href="https://huggingface.co/deepseek-ai/DeepSeek-V3.2-Exp">deepseek-ai/DeepSeek-V3.2-Exp</a>
      </td>
      <td className='modality'>text-to-text</td>
      <td className='encodings'>float8_e4m3fn</td>
      <td className='gpus'>Yes</td>
    </tr>
    <tr>
      <td className='arch'><a href="https://github.com/modular/modular/tree/main/max/python/max/pipelines/architectures/deepseekV3"><code>DeepseekV3ForCausalLM</code></a></td>
      <td className='models'>
        <a href="https://huggingface.co/deepseek-ai/DeepSeek-V3">deepseek-ai/DeepSeek-V3</a>
      </td>
      <td className='modality'>text-to-text</td>
      <td className='encodings'>bfloat16, float4_e2m1fnx2, float8_e4m3fn</td>
      <td className='gpus'>Yes</td>
    </tr>
    <tr>
      <td className='arch'><a href="https://github.com/modular/modular/tree/main/max/python/max/pipelines/architectures/dflash_llama3"><code>DFlashDraftModel</code></a></td>
      <td className='models'>
        <a href="https://huggingface.co/z-lab/LLaMA3.1-8B-Instruct-DFlash-UltraChat">z-lab/LLaMA3.1-8B-Instruct-DFlash-UltraChat</a>
      </td>
      <td className='modality'>text-to-text</td>
      <td className='encodings'>bfloat16, float32</td>
      <td className='gpus'>Yes</td>
    </tr>
    <tr>
      <td className='arch'><a href="https://github.com/modular/modular/tree/main/max/python/max/pipelines/architectures/exaone"><code>ExaoneForCausalLM</code></a></td>
      <td className='models'>
        <a href="https://huggingface.co/LGAI-EXAONE/EXAONE-3.5-2.4B-Instruct">LGAI-EXAONE/EXAONE-3.5-2.4B-Instruct</a>,<br/>
        <a href="https://huggingface.co/LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct">LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct</a>,<br/>
        <a href="https://huggingface.co/LGAI-EXAONE/EXAONE-3.5-32B-Instruct">LGAI-EXAONE/EXAONE-3.5-32B-Instruct</a>
      </td>
      <td className='modality'>text-to-text</td>
      <td className='encodings'>bfloat16, float32, q6_k</td>
      <td className='gpus'>No</td>
    </tr>
    <tr>
      <td className='arch'><a href="https://github.com/modular/modular/tree/main/max/python/max/pipelines/architectures/flux2"><code>Flux2KleinPipeline</code></a></td>
      <td className='models'>
        <a href="https://huggingface.co/black-forest-labs/FLUX.2-klein-4B">black-forest-labs/FLUX.2-klein-4B</a>,<br/>
        <a href="https://huggingface.co/black-forest-labs/FLUX.2-klein-9B">black-forest-labs/FLUX.2-klein-9B</a>,<br/>
        <a href="https://huggingface.co/black-forest-labs/FLUX.2-klein-base-4B">black-forest-labs/FLUX.2-klein-base-4B</a>,<br/>
        <a href="https://huggingface.co/black-forest-labs/FLUX.2-klein-base-9B">black-forest-labs/FLUX.2-klein-base-9B</a>,<br/>
        <a href="https://huggingface.co/black-forest-labs/FLUX.2-klein-4b-nvfp4">black-forest-labs/FLUX.2-klein-4b-nvfp4</a>,<br/>
        <a href="https://huggingface.co/black-forest-labs/FLUX.2-klein-9b-nvfp4">black-forest-labs/FLUX.2-klein-9b-nvfp4</a>
      </td>
      <td className='modality'>image-to-image,<br/>text-to-image</td>
      <td className='encodings'>bfloat16, float4_e2m1fnx2</td>
      <td className='gpus'>No</td>
    </tr>
    <tr>
      <td className='arch'><a href="https://github.com/modular/modular/tree/main/max/python/max/pipelines/architectures/flux2"><code>Flux2Pipeline</code></a></td>
      <td className='models'>
        <a href="https://huggingface.co/black-forest-labs/FLUX.2-dev">black-forest-labs/FLUX.2-dev</a>,<br/>
        <a href="https://huggingface.co/black-forest-labs/FLUX.2-dev-NVFP4">black-forest-labs/FLUX.2-dev-NVFP4</a>
      </td>
      <td className='modality'>image-to-image,<br/>text-to-image</td>
      <td className='encodings'>bfloat16, float4_e2m1fnx2</td>
      <td className='gpus'>No</td>
    </tr>
    <tr>
      <td className='arch'><a href="https://github.com/modular/modular/tree/main/max/python/max/pipelines/architectures/gemma3"><code>Gemma3ForCausalLM</code></a></td>
      <td className='models'>
        <a href="https://huggingface.co/google/gemma-3-1b-it">google/gemma-3-1b-it</a>,<br/>
        <a href="https://huggingface.co/google/gemma-3-1b-pt">google/gemma-3-1b-pt</a>
      </td>
      <td className='modality'>text-to-text</td>
      <td className='encodings'>bfloat16</td>
      <td className='gpus'>Yes</td>
    </tr>
    <tr>
      <td className='arch'><a href="https://github.com/modular/modular/tree/main/max/python/max/pipelines/architectures/gemma3multimodal"><code>Gemma3ForConditionalGeneration</code></a></td>
      <td className='models'>
        <a href="https://huggingface.co/google/gemma-3-4b-it">google/gemma-3-4b-it</a>,<br/>
        <a href="https://huggingface.co/google/gemma-3-4b-pt">google/gemma-3-4b-pt</a>,<br/>
        <a href="https://huggingface.co/google/gemma-3-12b-it">google/gemma-3-12b-it</a>,<br/>
        <a href="https://huggingface.co/google/gemma-3-12b-pt">google/gemma-3-12b-pt</a>,<br/>
        <a href="https://huggingface.co/google/gemma-3-27b-it">google/gemma-3-27b-it</a>,<br/>
        <a href="https://huggingface.co/google/gemma-3-27b-pt">google/gemma-3-27b-pt</a>
      </td>
      <td className='modality'>image-to-text,<br/>text-to-text</td>
      <td className='encodings'>bfloat16, float8_e4m3fn</td>
      <td className='gpus'>Yes</td>
    </tr>
    <tr>
      <td className='arch'><a href="https://github.com/modular/modular/tree/main/max/python/max/pipelines/architectures/gemma4"><code>Gemma4ForConditionalGeneration</code></a></td>
      <td className='models'>
        <a href="https://huggingface.co/google/gemma-4-31B-it">google/gemma-4-31B-it</a>,<br/>
        <a href="https://huggingface.co/nvidia/Gemma-4-31B-IT-NVFP4">nvidia/Gemma-4-31B-IT-NVFP4</a>
      </td>
      <td className='modality'>image-to-text,<br/>text-to-text,<br/>video-to-text</td>
      <td className='encodings'>bfloat16, float4_e2m1fnx2</td>
      <td className='gpus'>Yes</td>
    </tr>
    <tr>
      <td className='arch'><a href="https://github.com/modular/modular/tree/main/max/python/max/pipelines/architectures/glm5_1"><code>GlmMoeDsaForCausalLM</code></a></td>
      <td className='models'>
        <a href="https://huggingface.co/zai-org/GLM-5.1">zai-org/GLM-5.1</a>,<br/>
        <a href="https://huggingface.co/zai-org/GLM-5.1-FP8">zai-org/GLM-5.1-FP8</a>,<br/>
        <a href="https://huggingface.co/zai-org/GLM-5">zai-org/GLM-5</a>
      </td>
      <td className='modality'>text-to-text</td>
      <td className='encodings'>bfloat16, float4_e2m1fnx2, float8_e4m3fn</td>
      <td className='gpus'>Yes</td>
    </tr>
    <tr>
      <td className='arch'><a href="https://github.com/modular/modular/tree/main/max/python/max/pipelines/architectures/gpt_oss"><code>GptOssForCausalLM</code></a></td>
      <td className='models'>
        <a href="https://huggingface.co/openai/gpt-oss-20b">openai/gpt-oss-20b</a>,<br/>
        <a href="https://huggingface.co/openai/gpt-oss-120b">openai/gpt-oss-120b</a>,<br/>
        <a href="https://huggingface.co/unsloth/gpt-oss-20b-BF16">unsloth/gpt-oss-20b-BF16</a>
      </td>
      <td className='modality'>text-to-text</td>
      <td className='encodings'>bfloat16, float4_e2m1fnx2</td>
      <td className='gpus'>Yes</td>
    </tr>
    <tr>
      <td className='arch'><a href="https://github.com/modular/modular/tree/main/max/python/max/pipelines/architectures/granite"><code>GraniteForCausalLM</code></a></td>
      <td className='models'>
        <a href="https://huggingface.co/ibm-granite/granite-3.1-8b-instruct">ibm-granite/granite-3.1-8b-instruct</a>,<br/>
        <a href="https://huggingface.co/ibm-granite/granite-3.1-8b-base">ibm-granite/granite-3.1-8b-base</a>
      </td>
      <td className='modality'>text-to-text</td>
      <td className='encodings'>bfloat16, float32</td>
      <td className='gpus'>No</td>
    </tr>
    <tr>
      <td className='arch'><a href="https://github.com/modular/modular/tree/main/max/python/max/pipelines/architectures/hy_v3"><code>HYV3ForCausalLM</code></a></td>
      <td className='models'>
        <a href="https://huggingface.co/tencent/Hy3-preview">tencent/Hy3-preview</a>
      </td>
      <td className='modality'>text-to-text</td>
      <td className='encodings'>bfloat16</td>
      <td className='gpus'>Yes</td>
    </tr>
    <tr>
      <td className='arch'><a href="https://github.com/modular/modular/tree/main/max/python/max/pipelines/architectures/idefics3"><code>Idefics3ForConditionalGeneration</code></a></td>
      <td className='models'>
        <a href="https://huggingface.co/HuggingFaceM4/Idefics3-8B-Llama3">HuggingFaceM4/Idefics3-8B-Llama3</a>
      </td>
      <td className='modality'>image-to-text,<br/>text-to-text</td>
      <td className='encodings'>bfloat16</td>
      <td className='gpus'>No</td>
    </tr>
    <tr>
      <td className='arch'><a href="https://github.com/modular/modular/tree/main/max/python/max/pipelines/architectures/internvl"><code>InternVLChatModel</code></a></td>
      <td className='models'>
        <a href="https://huggingface.co/OpenGVLab/InternVL3-8B-Instruct">OpenGVLab/InternVL3-8B-Instruct</a>
      </td>
      <td className='modality'>image-to-text,<br/>text-to-text</td>
      <td className='encodings'>bfloat16</td>
      <td className='gpus'>Yes</td>
    </tr>
    <tr>
      <td className='arch'><a href="https://github.com/modular/modular/tree/main/max/python/max/pipelines/architectures/kimik2_5"><code>KimiK25ForConditionalGeneration</code></a></td>
      <td className='models'>
        <a href="https://huggingface.co/nvidia/Kimi-K2.5-NVFP4">nvidia/Kimi-K2.5-NVFP4</a>,<br/>
        <a href="https://huggingface.co/nvidia/Kimi-K2.6-NVFP4">nvidia/Kimi-K2.6-NVFP4</a>
      </td>
      <td className='modality'>image-to-text,<br/>text-to-text</td>
      <td className='encodings'>bfloat16, float4_e2m1fnx2, float8_e4m3fn</td>
      <td className='gpus'>Yes</td>
    </tr>
    <tr>
      <td className='arch'><a href="https://github.com/modular/modular/tree/main/max/python/max/pipelines/architectures/kimik2_5"><code>KimiVLForConditionalGeneration</code></a></td>
      <td className='models'>
        <a href="https://huggingface.co/moonshotai/Kimi-VL-A3B-Instruct">moonshotai/Kimi-VL-A3B-Instruct</a>
      </td>
      <td className='modality'>image-to-text,<br/>text-to-text</td>
      <td className='encodings'>bfloat16, float4_e2m1fnx2, float8_e4m3fn</td>
      <td className='gpus'>Yes</td>
    </tr>
    <tr>
      <td className='arch'><a href="https://github.com/modular/modular/tree/main/max/python/max/pipelines/architectures/lfm2"><code>Lfm2ForCausalLM</code></a></td>
      <td className='models'>
        <a href="https://huggingface.co/LiquidAI/LFM2.5-350M">LiquidAI/LFM2.5-350M</a>,<br/>
        <a href="https://huggingface.co/LiquidAI/LFM2.5-350M-Base">LiquidAI/LFM2.5-350M-Base</a>
      </td>
      <td className='modality'>text-to-text</td>
      <td className='encodings'>bfloat16, float32</td>
      <td className='gpus'>No</td>
    </tr>
    <tr>
      <td className='arch'><a href="https://github.com/modular/modular/tree/main/max/python/max/pipelines/architectures/llama3"><code>LlamaForCausalLM</code></a></td>
      <td className='models'>
        <a href="https://huggingface.co/meta-llama/Llama-3.1-8B-Instruct">meta-llama/Llama-3.1-8B-Instruct</a>,<br/>
        <a href="https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Llama-8B">deepseek-ai/DeepSeek-R1-Distill-Llama-8B</a>,<br/>
        <a href="https://huggingface.co/meta-llama/Llama-Guard-3-8B">meta-llama/Llama-Guard-3-8B</a>,<br/>
        <a href="https://huggingface.co/meta-llama/Llama-3.2-1B-Instruct">meta-llama/Llama-3.2-1B-Instruct</a>,<br/>
        <a href="https://huggingface.co/meta-llama/Llama-3.2-3B-Instruct">meta-llama/Llama-3.2-3B-Instruct</a>,<br/>
        <a href="https://huggingface.co/deepseek-ai/deepseek-coder-6.7b-instruct">deepseek-ai/deepseek-coder-6.7b-instruct</a>,<br/>
        <a href="https://huggingface.co/modularai/Llama-3.1-8B-Instruct-GGUF">modularai/Llama-3.1-8B-Instruct-GGUF</a>
      </td>
      <td className='modality'>text-to-text</td>
      <td className='encodings'>bfloat16, float32, float4_e2m1fnx2, float8_e4m3fn, gptq, q6_k</td>
      <td className='gpus'>Yes</td>
    </tr>
    <tr>
      <td className='arch'><a href="https://github.com/modular/modular/tree/main/max/python/max/pipelines/architectures/pixtral"><code>LlavaForConditionalGeneration</code></a></td>
      <td className='models'>
        <a href="https://huggingface.co/mistral-experimental/pixtral-12b">mistral-experimental/pixtral-12b</a>
      </td>
      <td className='modality'>image-to-text,<br/>text-to-text</td>
      <td className='encodings'>bfloat16</td>
      <td className='gpus'>No</td>
    </tr>
    <tr>
      <td className='arch'><a href="https://github.com/modular/modular/tree/main/max/python/max/pipelines/architectures/mamba"><code>MambaForCausalLM</code></a></td>
      <td className='models'>
        <a href="https://huggingface.co/state-spaces/mamba-130m-hf">state-spaces/mamba-130m-hf</a>
      </td>
      <td className='modality'>text-to-text</td>
      <td className='encodings'>bfloat16, float32</td>
      <td className='gpus'>No</td>
    </tr>
    <tr>
      <td className='arch'><a href="https://github.com/modular/modular/tree/main/max/python/max/pipelines/architectures/minimax_m2"><code>MiniMaxM2ForCausalLM</code></a></td>
      <td className='models'>
        <a href="https://huggingface.co/MiniMaxAI/MiniMax-M2.7">MiniMaxAI/MiniMax-M2.7</a>,<br/>
        <a href="https://huggingface.co/MiniMaxAI/MiniMax-M2.5">MiniMaxAI/MiniMax-M2.5</a>,<br/>
        <a href="https://huggingface.co/lukealonso/MiniMax-M2.7-NVFP4">lukealonso/MiniMax-M2.7-NVFP4</a>,<br/>
        <a href="https://huggingface.co/amd/MiniMax-M2.7-MXFP4">amd/MiniMax-M2.7-MXFP4</a>
      </td>
      <td className='modality'>text-to-text</td>
      <td className='encodings'>float4_e2m1fnx2, float8_e4m3fn</td>
      <td className='gpus'>Yes</td>
    </tr>
    <tr>
      <td className='arch'><a href="https://github.com/modular/modular/tree/main/max/python/max/pipelines/architectures/mistral3"><code>Mistral3ForConditionalGeneration</code></a></td>
      <td className='models'>
        <a href="https://huggingface.co/mistralai/Mistral-Small-3.1-24B-Instruct-2503">mistralai/Mistral-Small-3.1-24B-Instruct-2503</a>
      </td>
      <td className='modality'>text-to-text</td>
      <td className='encodings'>bfloat16</td>
      <td className='gpus'>Yes</td>
    </tr>
    <tr>
      <td className='arch'><a href="https://github.com/modular/modular/tree/main/max/python/max/pipelines/architectures/mistral"><code>MistralForCausalLM</code></a></td>
      <td className='models'>
        <a href="https://huggingface.co/mistralai/Mistral-Nemo-Instruct-2407">mistralai/Mistral-Nemo-Instruct-2407</a>
      </td>
      <td className='modality'>text-to-text</td>
      <td className='encodings'>bfloat16</td>
      <td className='gpus'>Yes</td>
    </tr>
    <tr>
      <td className='arch'><a href="https://github.com/modular/modular/tree/main/max/python/max/pipelines/architectures/mpnet"><code>MPNetForMaskedLM</code></a></td>
      <td className='models'>
        <a href="https://huggingface.co/sentence-transformers/all-mpnet-base-v2">sentence-transformers/all-mpnet-base-v2</a>
      </td>
      <td className='modality'>text-to-embeddings</td>
      <td className='encodings'>bfloat16, float32</td>
      <td className='gpus'>No</td>
    </tr>
    <tr>
      <td className='arch'><a href="https://github.com/modular/modular/tree/main/max/python/max/pipelines/architectures/olmo2"><code>Olmo2ForCausalLM</code></a></td>
      <td className='models'>
        <a href="https://huggingface.co/allenai/OLMo-2-0425-1B-Instruct">allenai/OLMo-2-0425-1B-Instruct</a>,<br/>
        <a href="https://huggingface.co/allenai/OLMo-2-1124-7B">allenai/OLMo-2-1124-7B</a>,<br/>
        <a href="https://huggingface.co/allenai/OLMo-2-1124-13B-Instruct">allenai/OLMo-2-1124-13B-Instruct</a>,<br/>
        <a href="https://huggingface.co/allenai/OLMo-2-0325-32B-Instruct">allenai/OLMo-2-0325-32B-Instruct</a>,<br/>
        <a href="https://huggingface.co/allenai/OLMo-2-1124-7B-GGUF">allenai/OLMo-2-1124-7B-GGUF</a>
      </td>
      <td className='modality'>text-to-text</td>
      <td className='encodings'>bfloat16, float32</td>
      <td className='gpus'>No</td>
    </tr>
    <tr>
      <td className='arch'><a href="https://github.com/modular/modular/tree/main/max/python/max/pipelines/architectures/olmo3"><code>Olmo3ForCausalLM</code></a></td>
      <td className='models'>
        <a href="https://huggingface.co/allenai/Olmo-3-7B-Instruct">allenai/Olmo-3-7B-Instruct</a>
      </td>
      <td className='modality'>text-to-text</td>
      <td className='encodings'>bfloat16</td>
      <td className='gpus'>No</td>
    </tr>
    <tr>
      <td className='arch'><a href="https://github.com/modular/modular/tree/main/max/python/max/pipelines/architectures/olmo"><code>OlmoForCausalLM</code></a></td>
      <td className='models'>
        <a href="https://huggingface.co/allenai/OLMo-1B-hf">allenai/OLMo-1B-hf</a>,<br/>
        <a href="https://huggingface.co/allenai/OLMo-1B-0724-hf">allenai/OLMo-1B-0724-hf</a>
      </td>
      <td className='modality'>text-to-text</td>
      <td className='encodings'>bfloat16, float32</td>
      <td className='gpus'>No</td>
    </tr>
    <tr>
      <td className='arch'><a href="https://github.com/modular/modular/tree/main/max/python/max/pipelines/architectures/phi3"><code>Phi3ForCausalLM</code></a></td>
      <td className='models'>
        <a href="https://huggingface.co/microsoft/phi-4">microsoft/phi-4</a>,<br/>
        <a href="https://huggingface.co/microsoft/Phi-3.5-mini-instruct">microsoft/Phi-3.5-mini-instruct</a>
      </td>
      <td className='modality'>text-to-text</td>
      <td className='encodings'>bfloat16, float32</td>
      <td className='gpus'>No</td>
    </tr>
    <tr>
      <td className='arch'><a href="https://github.com/modular/modular/tree/main/max/python/max/pipelines/architectures/qwen2_5vl"><code>Qwen2_5_VLForConditionalGeneration</code></a></td>
      <td className='models'>
        <a href="https://huggingface.co/Qwen/Qwen2.5-VL-3B-Instruct">Qwen/Qwen2.5-VL-3B-Instruct</a>,<br/>
        <a href="https://huggingface.co/Qwen/Qwen2.5-VL-7B-Instruct">Qwen/Qwen2.5-VL-7B-Instruct</a>
      </td>
      <td className='modality'>image-to-text,<br/>text-to-text</td>
      <td className='encodings'>bfloat16, float32, float8_e4m3fn</td>
      <td className='gpus'>Yes</td>
    </tr>
    <tr>
      <td className='arch'><a href="https://github.com/modular/modular/tree/main/max/python/max/pipelines/architectures/qwen2"><code>Qwen2ForCausalLM</code></a></td>
      <td className='models'>
        <a href="https://huggingface.co/Qwen/Qwen2.5-7B-Instruct">Qwen/Qwen2.5-7B-Instruct</a>,<br/>
        <a href="https://huggingface.co/Qwen/QwQ-32B">Qwen/QwQ-32B</a>
      </td>
      <td className='modality'>text-to-text</td>
      <td className='encodings'>bfloat16, float32</td>
      <td className='gpus'>Yes</td>
    </tr>
    <tr>
      <td className='arch'><a href="https://github.com/modular/modular/tree/main/max/python/max/pipelines/architectures/qwen3_5"><code>Qwen3_5ForConditionalGeneration</code></a></td>
      <td className='models'>
        <a href="https://huggingface.co/Qwen/Qwen3.5-27B">Qwen/Qwen3.5-27B</a>
      </td>
      <td className='modality'>text-to-text</td>
      <td className='encodings'>bfloat16, float32</td>
      <td className='gpus'>No</td>
    </tr>
    <tr>
      <td className='arch'><a href="https://github.com/modular/modular/tree/main/max/python/max/pipelines/architectures/qwen3"><code>Qwen3ForCausalLM</code></a></td>
      <td className='models'>
        <a href="https://huggingface.co/Qwen/Qwen3-8B">Qwen/Qwen3-8B</a>,<br/>
        <a href="https://huggingface.co/Qwen/Qwen3-30B-A3B">Qwen/Qwen3-30B-A3B</a>,<br/>
        <a href="https://huggingface.co/Qwen/Qwen3-Embedding-0.6B">Qwen/Qwen3-Embedding-0.6B</a>,<br/>
        <a href="https://huggingface.co/Qwen/Qwen3-Embedding-4B">Qwen/Qwen3-Embedding-4B</a>,<br/>
        <a href="https://huggingface.co/Qwen/Qwen3-Embedding-8B">Qwen/Qwen3-Embedding-8B</a>
      </td>
      <td className='modality'>text-to-embeddings,<br/>text-to-text</td>
      <td className='encodings'>bfloat16, float32, float8_e4m3fn</td>
      <td className='gpus'>Yes</td>
    </tr>
    <tr>
      <td className='arch'><a href="https://github.com/modular/modular/tree/main/max/python/max/pipelines/architectures/qwen3"><code>Qwen3MoeForCausalLM</code></a></td>
      <td className='models'>
        <a href="https://huggingface.co/Qwen/Qwen3-30B-A3B-Instruct">Qwen/Qwen3-30B-A3B-Instruct</a>,<br/>
        <a href="https://huggingface.co/Qwen/Qwen3-30B-A3B-Instruct-2507-FP8">Qwen/Qwen3-30B-A3B-Instruct-2507-FP8</a>
      </td>
      <td className='modality'>text-to-text</td>
      <td className='encodings'>bfloat16, float32, float8_e4m3fn</td>
      <td className='gpus'>Yes</td>
    </tr>
    <tr>
      <td className='arch'><a href="https://github.com/modular/modular/tree/main/max/python/max/pipelines/architectures/qwen3vl_moe"><code>Qwen3VLForConditionalGeneration</code></a></td>
      <td className='models'>
        <a href="https://huggingface.co/Qwen/Qwen3-VL-4B-Instruct">Qwen/Qwen3-VL-4B-Instruct</a>,<br/>
        <a href="https://huggingface.co/Qwen/Qwen3-VL-2B-Instruct">Qwen/Qwen3-VL-2B-Instruct</a>
      </td>
      <td className='modality'>image-to-text,<br/>text-to-text</td>
      <td className='encodings'>bfloat16, float32, float8_e4m3fn</td>
      <td className='gpus'>Yes</td>
    </tr>
    <tr>
      <td className='arch'><a href="https://github.com/modular/modular/tree/main/max/python/max/pipelines/architectures/qwen3vl_moe"><code>Qwen3VLMoeForConditionalGeneration</code></a></td>
      <td className='models'>
        <a href="https://huggingface.co/Qwen/Qwen3-VL-30B-A3B-Instruct">Qwen/Qwen3-VL-30B-A3B-Instruct</a>
      </td>
      <td className='modality'>image-to-text,<br/>text-to-text</td>
      <td className='encodings'>bfloat16, float32, float8_e4m3fn</td>
      <td className='gpus'>Yes</td>
    </tr>
    <tr>
      <td className='arch'><a href="https://github.com/modular/modular/tree/main/max/python/max/pipelines/architectures/qwen_image_edit"><code>QwenImageEditPipeline</code></a></td>
      <td className='models'>
        <a href="https://huggingface.co/Qwen/Qwen-Image-Edit-2511">Qwen/Qwen-Image-Edit-2511</a>
      </td>
      <td className='modality'>image-to-image,<br/>text-to-image</td>
      <td className='encodings'>bfloat16</td>
      <td className='gpus'>No</td>
    </tr>
    <tr>
      <td className='arch'><a href="https://github.com/modular/modular/tree/main/max/python/max/pipelines/architectures/qwen_image_edit"><code>QwenImageEditPlusPipeline</code></a></td>
      <td className='models'>
        <a href="https://huggingface.co/Qwen/Qwen-Image-Edit-2511">Qwen/Qwen-Image-Edit-2511</a>
      </td>
      <td className='modality'>image-to-image,<br/>text-to-image</td>
      <td className='encodings'>bfloat16</td>
      <td className='gpus'>No</td>
    </tr>
    <tr>
      <td className='arch'><a href="https://github.com/modular/modular/tree/main/max/python/max/pipelines/architectures/qwen_image"><code>QwenImagePipeline</code></a></td>
      <td className='models'>
        <a href="https://huggingface.co/Qwen/Qwen-Image-2512">Qwen/Qwen-Image-2512</a>
      </td>
      <td className='modality'>text-to-image</td>
      <td className='encodings'>bfloat16</td>
      <td className='gpus'>No</td>
    </tr>
    <tr>
      <td className='arch'><a href="https://github.com/modular/modular/tree/main/max/python/max/pipelines/architectures/step3p5"><code>Step3p5ForCausalLM</code></a></td>
      <td className='models'>
        <a href="https://huggingface.co/stepfun-ai/Step-3.5-Flash">stepfun-ai/Step-3.5-Flash</a>
      </td>
      <td className='modality'>text-to-text</td>
      <td className='encodings'>bfloat16</td>
      <td className='gpus'>Yes</td>
    </tr>
    <tr>
      <td className='arch'><a href="https://github.com/modular/modular/tree/main/max/python/max/pipelines/architectures/unified_dflash_kimi_k25"><code>UnifiedDflashKimiK25ForCausalLM</code></a></td>
      <td className='models'>
        <a href="https://huggingface.co/nvidia/Kimi-K2.5-NVFP4">nvidia/Kimi-K2.5-NVFP4</a>
      </td>
      <td className='modality'>image-to-text,<br/>text-to-text</td>
      <td className='encodings'>bfloat16, float4_e2m1fnx2, float8_e4m3fn</td>
      <td className='gpus'>Yes</td>
    </tr>
    <tr>
      <td className='arch'><a href="https://github.com/modular/modular/tree/main/max/python/max/pipelines/architectures/unified_dflash_llama3"><code>UnifiedDflashLlama3ForCausalLM</code></a></td>
      <td className='models'>
        <a href="https://huggingface.co/meta-llama/Llama-3.2-3B-Instruct">meta-llama/Llama-3.2-3B-Instruct</a>
      </td>
      <td className='modality'>text-to-text</td>
      <td className='encodings'>bfloat16, float32</td>
      <td className='gpus'>No</td>
    </tr>
    <tr>
      <td className='arch'><a href="https://github.com/modular/modular/tree/main/max/python/max/pipelines/architectures/unified_mtp_deepseekV3"><code>UnifiedMTPDeepseekV3ForCausalLM</code></a></td>
      <td className='models'>
        <a href="https://huggingface.co/deepseek-ai/DeepSeek-V3">deepseek-ai/DeepSeek-V3</a>
      </td>
      <td className='modality'>text-to-text</td>
      <td className='encodings'>bfloat16, float4_e2m1fnx2, float8_e4m3fn</td>
      <td className='gpus'>Yes</td>
    </tr>
    <tr>
      <td className='arch'><a href="https://github.com/modular/modular/tree/main/max/python/max/pipelines/architectures/wan"><code>WanImageToVideoPipeline</code></a></td>
      <td className='models'>
        <a href="https://huggingface.co/Wan-AI/Wan2.2-I2V-A14B-Diffusers">Wan-AI/Wan2.2-I2V-A14B-Diffusers</a>,<br/>
        <a href="https://huggingface.co/Wan-AI/Wan2.1-I2V-14B-720P-Diffusers">Wan-AI/Wan2.1-I2V-14B-720P-Diffusers</a>
      </td>
      <td className='modality'>text-to-image</td>
      <td className='encodings'>bfloat16, float32</td>
      <td className='gpus'>No</td>
    </tr>
    <tr>
      <td className='arch'><a href="https://github.com/modular/modular/tree/main/max/python/max/pipelines/architectures/wan"><code>WanPipeline</code></a></td>
      <td className='models'>
        <a href="https://huggingface.co/Wan-AI/Wan2.2-T2V-A14B-Diffusers">Wan-AI/Wan2.2-T2V-A14B-Diffusers</a>,<br/>
        <a href="https://huggingface.co/Wan-AI/Wan2.1-T2V-14B-Diffusers">Wan-AI/Wan2.1-T2V-14B-Diffusers</a>,<br/>
        <a href="https://huggingface.co/Wan-AI/Wan2.2-TI2V-5B-Diffusers">Wan-AI/Wan2.2-TI2V-5B-Diffusers</a>,<br/>
        <a href="https://huggingface.co/yetter-ai/Wan2.2-TI2V-5B-Turbo-Diffusers">yetter-ai/Wan2.2-TI2V-5B-Turbo-Diffusers</a>
      </td>
      <td className='modality'>text-to-image</td>
      <td className='encodings'>bfloat16, float32</td>
      <td className='gpus'>No</td>
    </tr>
    <tr>
      <td className='arch'><a href="https://github.com/modular/modular/tree/main/max/python/max/pipelines/architectures/z_image_modulev3"><code>ZImagePipeline</code></a></td>
      <td className='models'>
        <a href="https://huggingface.co/Tongyi-MAI/Z-Image">Tongyi-MAI/Z-Image</a>,<br/>
        <a href="https://huggingface.co/Zyphra/Z-Image">Zyphra/Z-Image</a>
      </td>
      <td className='modality'>text-to-image</td>
      <td className='encodings'>bfloat16</td>
      <td className='gpus'>No</td>
    </tr>
  </tbody>
</table>