The tables on this page are generated from Dew’s source at build time. Pretrained.load(source) reads a Hugging Face directory or Hub repository whose config.json names one of the model_type values below, and dew.pipeline(source) wraps the same load in a generation task. Pretrained.save writes trained weights back in the source’s own layout. Transformers loads the exported directory for the registered causal decoder families; a text-only DiffusionGemma export is refused before anything is written.
A port counts as supported when a test loads the same weights into Dew and into the reference implementation and compares the outputs in float32. For most families that test runs on a small fixture with the release’s own configuration and tensor shapes. The “Checked with” column says when a family was also compared at full size against a released checkpoint. Most families have not been run at full size; memory use and throughput at full size are unmeasured for them.
A checkpoint whose type is not listed can still load. A type that computes a registered family’s block is verified against transformers at load time. Any other causal LM that Transformers can build loads through Transformers’ own PyTorch forward, lowered to JAX by torchax, its attention Dew’s (packing included) wherever its layers ask Transformers’ attention interface for it, generating through Dew’s decode cache where its attention is Dew’s, with no sharding rules, and fallback="torchax" takes that route for any checkpoint. GGUF files, pytorch_model.bin repositories and mamba_ssm checkpoints load too. Train language models explains each of these paths and what it costs.
These families were asked for and are not in the tables above. Each one is tracked until it loads and a test compares it with its reference:
| Family | What is missing |
|---|
| MiniMax M3 | The shared expert, MTP and vision-language half, and a reader. |
| Kimi K2.5 and K3 vision | The vision towers and projectors. Both load as text decoders. |
| MiniMax H3 | Its configuration, to check whether the M2 reader covers it. |
| Family | model_type | Architecture | Checked with |
|---|
| GPT-2 | gpt2 | GPT2LMHeadModel | A fixture shaped like the release |
| GPT-NeoX | gpt_neox | GPTNeoXForCausalLM | A fixture shaped like the release |
| OPT | opt | OPTForCausalLM | A fixture shaped like the release |
| GPT-Neo | gpt_neo | GPTNeoForCausalLM | A fixture shaped like the release |
| GPT-1 | openai-gpt | OpenAIGPTLMHeadModel | A fixture shaped like the release |
| Phi | phi | PhiForCausalLM | A fixture shaped like the release |
| Phi-3 | phi3 | Phi3ForCausalLM | A fixture shaped like the release |
| Falcon | falcon | FalconForCausalLM | A fixture shaped like the release |
| GPT-J | gptj | GPTJForCausalLM | A fixture shaped like the release |
| GPTBigCode | gpt_bigcode | GPTBigCodeForCausalLM | A fixture shaped like the release |
| StarCoder2 | starcoder2 | Starcoder2ForCausalLM | A fixture shaped like the release |
| Llama | llama | LlamaForCausalLM | SmolLM2-135M, logits against transformers; its GGUF Q8_0 and Q4_K_M files |
| Mistral | mistral | MistralForCausalLM | A fixture shaped like the release |
| Ministral | ministral | MinistralForCausalLM | A fixture shaped like the release |
| Qwen 2 | qwen2 | Qwen2ForCausalLM | A fixture shaped like the release |
| Qwen 3 | qwen3 | Qwen3ForCausalLM | Qwen3-0.6B, logits against transformers, on one GPU, and streamed onto an 8-device CPU mesh |
| Ouro (looped) | ouro | OuroForCausalLM | A fixture shaped like the release |
| Dory (looped nGPT) | dory | DoryForCausalLM | A fixture shaped like the release |
| Qwen 3.5, text | qwen3_5_text | Qwen3_5ForCausalLM | A fixture shaped like the release |
| K2 Horizon | k2_horizon | K2HorizonForCausalLM | A fixture shaped like the release |
| Muse Glimmer, text | muse_glimmer_text | MuseGlimmerTextModel | A fixture shaped like the release |
| BLOOM | bloom | BloomForCausalLM | A fixture shaped like the release |
| Gemma | gemma | GemmaForCausalLM | A fixture shaped like the release |
| Gemma 2 | gemma2 | Gemma2ForCausalLM | A fixture shaped like the release |
| Gemma 3, text | gemma3_text | Gemma3ForCausalLM | A fixture shaped like the release |
| Gemma 3n, text | gemma3n_text | Gemma3nForCausalLM | A fixture shaped like the release |
| Gemma 4, text | gemma4_text | Gemma4ForCausalLM | A fixture shaped like the release |
| Gemma 4 Unified, text | gemma4_unified_text | Gemma4UnifiedForCausalLM | A fixture shaped like the release |
| OLMo 3 | olmo3 | Olmo3ForCausalLM | A fixture shaped like the release |
| Family | model_type | Architecture | Checked with |
|---|
| MiMo-V2 | mimo_v2 | MiMoV2ForCausalLM | A fixture shaped like the release |
| Mixtral | mixtral | MixtralForCausalLM | A fixture shaped like the release |
| MiniMax-M2 | minimax_m2 | MiniMaxM2ForCausalLM | A fixture shaped like the release |
| MiMo-V2-Flash | mimo_v2_flash | MiMoV2FlashForCausalLM | A fixture shaped like the release |
| Granite MoE | granitemoe | GraniteMoeForCausalLM | A fixture shaped like the release |
| Qwen3-MoE | qwen3_moe | Qwen3MoeForCausalLM | A fixture shaped like the release |
| Qwen 2 MoE | qwen2_moe | Qwen2MoeForCausalLM | A fixture shaped like the release |
| Qwen 3.5 MoE, text | qwen3_5_moe_text | Qwen3_5MoeForCausalLM | A fixture shaped like the release |
| Qwen3-VL-MoE, text | qwen3_vl_moe_text | Qwen3MoeForCausalLM | A fixture shaped like the release |
| gpt-oss | gpt_oss | GptOssForCausalLM | A fixture shaped like the release |
| Llama 4, text | llama4_text | Llama4ForCausalLM | A fixture shaped like the release |
| GLM 4 MoE | glm4_moe | Glm4MoeForCausalLM | A fixture shaped like the release |
| GLM-5.3 | glm_moe_dsa | GlmMoeDsaForCausalLM | A fixture shaped like the release |
| DeepSeek V2 | deepseek_v2 | DeepseekV2ForCausalLM | A fixture shaped like the release |
| DeepSeek V3 | deepseek_v3 | DeepseekV3ForCausalLM | A fixture shaped like the release |
| DeepSeek V3.2 | deepseek_v32 | DeepseekV32ForCausalLM | A fixture shaped like the release |
| DeepSeek V4 | deepseek_v4 | DeepseekV4ForCausalLM | A fixture shaped like the release |
| DeepSeek V4.1 | deepseek_v41 | DeepseekV41ForCausalLM | A fixture shaped like the release |
| Kimi K2 | kimi_k2 | DeepseekV3ForCausalLM | A fixture shaped like the release |
| Kimi K2.5, text | kimi_k25 | Kimi_K25ForConditionalGeneration | A fixture shaped like the release |
| Family | model_type | Architecture | Checked with |
|---|
| Qwen3-Next | qwen3_next | Qwen3NextForCausalLM | A fixture shaped like the release |
| Nemotron-H | nemotron_h | NemotronHForCausalLM | A fixture shaped like the release |
| GLM-5.3-Flash, text | glm5_next_text | Glm5NextTextForCausalLM | A fixture shaped like the release |
| Kimi Linear | kimi_linear | KimiLinearForCausalLM | A fixture shaped like the release |
| LFM3 | lfm3 | Lfm3ForCausalLM | A fixture shaped like the release |
| Kimi K3, text | kimi_k3 | KimiK3ForConditionalGeneration | A fixture shaped like the release |
| MiniMax-Text-01 and MiniMax-M1 | minimax, minimax_text_01, minimax_m1 | MiniMaxForCausalLM | A fixture shaped like the release |
| Qwen4-Exp (Qwen3.8-Flash-Next), text | qwen4_exp_text | Qwen4ExpForCausalLM | A fixture shaped like the release |
| Mamba | mamba | MambaForCausalLM | A fixture shaped like the release |
| Mamba-2 | mamba2 | Mamba2ForCausalLM | state-spaces/mamba2-130m against AntonV/mamba2-130m-hf |
| Family | model_type | Architecture | Checked with |
|---|
| LLaDA | llada | LLaDAModelLM | A fixture shaped like the release |
| LLaDA-MoE v2 | llada_moe_v2 | LLaDAMoEv2ModelLM | A fixture shaped like the release |
| LLaDA2 MoE | llada2_moe | LLaDA2MoeModelLM | A fixture shaped like the release |
| Dream | dream, Dream | DreamModel | A fixture shaped like the release |
| Diffusion Gemma | diffusion_gemma (text config diffusion_gemma_text) | DiffusionGemmaForBlockDiffusion | A fixture shaped like the release |
| Family | model_type | Architecture | Checked with |
|---|
| ModernBERT | modernbert | ModernBertForMaskedLM | A fixture shaped like the release |
These load the whole checkpoint: the text decoder listed above, the media towers and the checkpoint’s own processor, but where the media column says the language model alone, whose towers have no port yet.
| Family | model_type | Media |
|---|
| Gemma 3 | gemma3 | Images |
| Llama 4 | llama4 | Images |
| Gemma 4 | gemma4 | Images, video, audio |
| Gemma 4 Unified | gemma4_unified | Images, video, audio |
| Qwen 3.5 | qwen3_5 | Images, video |
| Qwen 3.5 MoE | qwen3_5_moe | Images, video |
| Qwen4-Exp (Qwen3.8-Flash-Next) | qwen4_exp | Images, video |
| Gemma 3n | gemma3n | Images, audio |
| Muse Glimmer | muse_glimmer | Images |
| LFM3-VL | lfm3_vl | Images, video |
| Qwen2-VL | qwen2_vl | None yet: its language model alone (text_only) |
| Qwen2.5-VL | qwen2_5_vl | None yet: its language model alone (text_only) |
| Qwen3-VL | qwen3_vl | None yet: its language model alone (text_only) |
| Qwen3-VL-MoE | qwen3_vl_moe | None yet: its language model alone (text_only) |
| MiMo-V2.6 | mimo_v2 | Images, video, audio |
| DeepSeek V4.1 | deepseek_v41 | Images |
| Kimi K2.5 | kimi_k25 | Images, video |
| Kimi K3 | kimi_k3 | Images, video |
Pretrained.load and dew.pipeline read a diffusers pipeline directory by the class it names in model_index.json.
| Family | Pipeline class | Task |
|---|
| Stable Diffusion | StableDiffusionPipeline | Text to image |
| Stable Diffusion | StableDiffusionImg2ImgPipeline | Image to image |
| Stable Diffusion | StableDiffusionInpaintPipeline | Inpainting |
| Stable Diffusion XL | StableDiffusionXLPipeline | Text to image |
| Stable Diffusion XL | StableDiffusionXLImg2ImgPipeline | Image to image, refiner |
| Stable Diffusion XL | StableDiffusionXLInpaintPipeline | Inpainting |
| Stable Diffusion 3 | StableDiffusion3Pipeline | Text to image |
| Flux | FluxPipeline | Text to image |
| Qwen-Image 2.1 | QwenImage21Pipeline | Text to image |
| FLUX.2 [dev] | Flux2Pipeline | Text to image |
| FLUX.2 [klein] | Flux2KleinPipeline | Text to image |
| Z-Image | ZImagePipeline | Text to image |
| Wan 2.1 | WanPipeline | Text to video |
| Stable Diffusion, Flax weights | FlaxStableDiffusionPipeline | Text to image |
| Stable Diffusion, Flax weights | FlaxStableDiffusionImg2ImgPipeline | Image to image |
| Stable Diffusion, Flax weights | FlaxStableDiffusionInpaintPipeline | Inpainting |
| Stable Diffusion XL, Flax weights | FlaxStableDiffusionXLPipeline | Text to image |
Code builds each one from its class; a run record names it by its registered name. A class the API reference documents links to its entry.