Upload folder using huggingface_hub

.gitattributes

@@ -33,3 +33,5 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+processor/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+tokenizer/tokenizer.json filter=lfs diff=lfs merge=lfs -text

audio_scheduler/scheduler_config.json

@@ -0,0 +1,5 @@
+{
+  "_class_name": "MiniMaxH3Scheduler",
+  "_diffusers_version": "0.40.0.dev0",
+  "shift": 3.0
+}

audio_vae/config.json

@@ -0,0 +1,56 @@
+{
+  "_class_name": "AutoencoderKLMiniMaxH3Audio",
+  "_diffusers_version": "0.40.0.dev0",
+  "decoder_dim": 16,
+  "decoder_kernel_sizes": [
+    4,
+    4
+  ],
+  "decoder_rates": [
+    2,
+    2
+  ],
+  "encoder_dim": 4,
+  "encoder_rates": [
+    2,
+    2
+  ],
+  "latent_channels": 8,
+  "latent_dim": 32,
+  "latents_mean": [
+    0.0,
+    0.0,
+    0.0,
+    0.0,
+    0.0,
+    0.0,
+    0.0,
+    0.0
+  ],
+  "latents_std": [
+    1.0,
+    1.0,
+    1.0,
+    1.0,
+    1.0,
+    1.0,
+    1.0,
+    1.0
+  ],
+  "num_attention_heads": 2,
+  "resblock_dilation_sizes": [
+    [
+      1,
+      3
+    ],
+    [
+      1,
+      3
+    ]
+  ],
+  "resblock_kernel_sizes": [
+    3,
+    7
+  ],
+  "sampling_rate": 160
+}

audio_vae/diffusion_pytorch_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5b713e1a4d05d24615a9675c67c91df06f0db7e0c28c583ec0df595b5353a405
+size 91020

model_index.json

@@ -0,0 +1,40 @@
+{
+  "_class_name": "MiniMaxH3Pipeline",
+  "_diffusers_version": "0.40.0.dev0",
+  "text_encoder": [
+    "transformers",
+    "Qwen3VLForConditionalGeneration"
+  ],
+  "tokenizer": [
+    "transformers",
+    "Qwen2Tokenizer"
+  ],
+  "processor": [
+    "transformers",
+    "Qwen3VLProcessor"
+  ],
+  "vae": [
+    "diffusers",
+    "AutoencoderKLMiniMaxH3"
+  ],
+  "audio_vae": [
+    "diffusers",
+    "AutoencoderKLMiniMaxH3Audio"
+  ],
+  "transformer": [
+    "diffusers",
+    "MiniMaxH3Transformer3DModel"
+  ],
+  "transformer_ref": [
+    "diffusers",
+    "MiniMaxH3Transformer3DModel"
+  ],
+  "scheduler": [
+    "diffusers",
+    "MiniMaxH3Scheduler"
+  ],
+  "audio_scheduler": [
+    "diffusers",
+    "MiniMaxH3Scheduler"
+  ]
+}

modular_model_index.json

@@ -0,0 +1,131 @@
+{
+  "_class_name": "MiniMaxH3ModularPipeline",
+  "_diffusers_version": "0.40.0.dev0",
+  "_blocks_class_name": "MiniMaxH3Blocks",
+  "text_encoder": [
+    "transformers",
+    "Qwen3VLForConditionalGeneration",
+    {
+      "type_hint": [
+        "transformers",
+        "Qwen3VLForConditionalGeneration"
+      ],
+      "pretrained_model_name_or_path": "diffusers-internal-dev/tiny-minimax-h3-modular-pipe",
+      "subfolder": "text_encoder",
+      "variant": null,
+      "revision": null
+    }
+  ],
+  "tokenizer": [
+    "transformers",
+    "Qwen2Tokenizer",
+    {
+      "type_hint": [
+        "transformers",
+        "Qwen2Tokenizer"
+      ],
+      "pretrained_model_name_or_path": "diffusers-internal-dev/tiny-minimax-h3-modular-pipe",
+      "subfolder": "tokenizer",
+      "variant": null,
+      "revision": null
+    }
+  ],
+  "processor": [
+    "transformers",
+    "Qwen3VLProcessor",
+    {
+      "type_hint": [
+        "transformers",
+        "Qwen3VLProcessor"
+      ],
+      "pretrained_model_name_or_path": "diffusers-internal-dev/tiny-minimax-h3-modular-pipe",
+      "subfolder": "processor",
+      "variant": null,
+      "revision": null
+    }
+  ],
+  "vae": [
+    "diffusers",
+    "AutoencoderKLMiniMaxH3",
+    {
+      "type_hint": [
+        "diffusers",
+        "AutoencoderKLMiniMaxH3"
+      ],
+      "pretrained_model_name_or_path": "diffusers-internal-dev/tiny-minimax-h3-modular-pipe",
+      "subfolder": "vae",
+      "variant": null,
+      "revision": null
+    }
+  ],
+  "audio_vae": [
+    "diffusers",
+    "AutoencoderKLMiniMaxH3Audio",
+    {
+      "type_hint": [
+        "diffusers",
+        "AutoencoderKLMiniMaxH3Audio"
+      ],
+      "pretrained_model_name_or_path": "diffusers-internal-dev/tiny-minimax-h3-modular-pipe",
+      "subfolder": "audio_vae",
+      "variant": null,
+      "revision": null
+    }
+  ],
+  "transformer": [
+    "diffusers",
+    "MiniMaxH3Transformer3DModel",
+    {
+      "type_hint": [
+        "diffusers",
+        "MiniMaxH3Transformer3DModel"
+      ],
+      "pretrained_model_name_or_path": "diffusers-internal-dev/tiny-minimax-h3-modular-pipe",
+      "subfolder": "transformer",
+      "variant": null,
+      "revision": null
+    }
+  ],
+  "transformer_ref": [
+    "diffusers",
+    "MiniMaxH3Transformer3DModel",
+    {
+      "type_hint": [
+        "diffusers",
+        "MiniMaxH3Transformer3DModel"
+      ],
+      "pretrained_model_name_or_path": "diffusers-internal-dev/tiny-minimax-h3-modular-pipe",
+      "subfolder": "transformer_ref",
+      "variant": null,
+      "revision": null
+    }
+  ],
+  "scheduler": [
+    "diffusers",
+    "MiniMaxH3Scheduler",
+    {
+      "type_hint": [
+        "diffusers",
+        "MiniMaxH3Scheduler"
+      ],
+      "pretrained_model_name_or_path": "diffusers-internal-dev/tiny-minimax-h3-modular-pipe",
+      "subfolder": "scheduler",
+      "variant": null,
+      "revision": null
+    }
+  ],
+  "audio_scheduler": [
+    "diffusers",
+    "MiniMaxH3Scheduler",
+    {
+      "type_hint": [
+        "diffusers",
+        "MiniMaxH3Scheduler"
+      ],
+      "pretrained_model_name_or_path": "diffusers-internal-dev/tiny-minimax-h3-modular-pipe",
+      "subfolder": "audio_scheduler",
+      "variant": null,
+      "revision": null
+    }
+  ]
+}

processor/chat_template.jinja

@@ -0,0 +1,7 @@
+{% set image_count = namespace(value=0) %}{% set video_count = namespace(value=0) %}{% for message in messages %}{% if loop.first and message['role'] != 'system' %}<|im_start|>system
+You are a helpful assistant.<|im_end|>
+{% endif %}<|im_start|>{{ message['role'] }}
+{% if message['content'] is string %}{{ message['content'] }}<|im_end|>
+{% else %}{% for content in message['content'] %}{% if content['type'] == 'image' or 'image' in content or 'image_url' in content %}{% set image_count.value = image_count.value + 1 %}{% if add_vision_id %}Picture {{ image_count.value }}: {% endif %}<|vision_start|><|image_pad|><|vision_end|>{% elif content['type'] == 'video' or 'video' in content %}{% set video_count.value = video_count.value + 1 %}{% if add_vision_id %}Video {{ video_count.value }}: {% endif %}<|vision_start|><|video_pad|><|vision_end|>{% elif 'text' in content %}{{ content['text'] }}{% endif %}{% endfor %}<|im_end|>
+{% endif %}{% endfor %}{% if add_generation_prompt %}<|im_start|>assistant
+{% endif %}

processor/processor_config.json

@@ -0,0 +1,62 @@
+{
+  "image_processor": {
+    "do_convert_rgb": true,
+    "do_normalize": true,
+    "do_rescale": true,
+    "do_resize": true,
+    "image_mean": [
+      0.48145466,
+      0.4578275,
+      0.40821073
+    ],
+    "image_processor_type": "Qwen2VLImageProcessor",
+    "image_std": [
+      0.26862954,
+      0.26130258,
+      0.27577711
+    ],
+    "merge_size": 2,
+    "patch_size": 4,
+    "resample": 3,
+    "rescale_factor": 0.00392156862745098,
+    "size": {
+      "longest_edge": 256,
+      "shortest_edge": 64
+    },
+    "temporal_patch_size": 2
+  },
+  "processor_class": "Qwen3VLProcessor",
+  "video_processor": {
+    "do_convert_rgb": true,
+    "do_normalize": true,
+    "do_rescale": true,
+    "do_resize": true,
+    "do_sample_frames": true,
+    "fps": 2,
+    "image_mean": [
+      0.5,
+      0.5,
+      0.5
+    ],
+    "image_std": [
+      0.5,
+      0.5,
+      0.5
+    ],
+    "max_frames": 768,
+    "max_pixels": 256,
+    "merge_size": 2,
+    "min_frames": 4,
+    "min_pixels": 64,
+    "patch_size": 4,
+    "resample": 3,
+    "rescale_factor": 0.00392156862745098,
+    "return_metadata": false,
+    "size": {
+      "longest_edge": 786432,
+      "shortest_edge": 131072
+    },
+    "temporal_patch_size": 2,
+    "video_processor_type": "Qwen3VLVideoProcessor"
+  }
+}

processor/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:312c03cb421e18f41c20c3e35f3302479a3a8f8b1f06779fd900326a0d32c22f
+size 11420534

processor/tokenizer_config.json

@@ -0,0 +1,35 @@
+{
+  "add_prefix_space": false,
+  "backend": "tokenizers",
+  "bos_token": null,
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "errors": "replace",
+  "extra_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "is_local": false,
+  "local_files_only": false,
+  "max_length": null,
+  "model_max_length": 32768,
+  "pad_to_multiple_of": null,
+  "pad_token": "<|endoftext|>",
+  "pad_token_type_id": 0,
+  "padding_side": "left",
+  "processor_class": "Qwen3VLProcessor",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

scheduler/scheduler_config.json

@@ -0,0 +1,5 @@
+{
+  "_class_name": "MiniMaxH3Scheduler",
+  "_diffusers_version": "0.40.0.dev0",
+  "shift": 12.0
+}

text_encoder/config.json

@@ -0,0 +1,59 @@
+{
+  "architectures": [
+    "Qwen3VLForConditionalGeneration"
+  ],
+  "dtype": "float32",
+  "image_token_id": 151655,
+  "model_type": "qwen3_vl",
+  "text_config": {
+    "attention_bias": false,
+    "attention_dropout": 0.0,
+    "head_dim": 8,
+    "hidden_act": "silu",
+    "hidden_size": 16,
+    "initializer_range": 0.02,
+    "intermediate_size": 16,
+    "max_position_embeddings": 128000,
+    "model_type": "qwen3_vl_text",
+    "num_attention_heads": 2,
+    "num_hidden_layers": 51,
+    "num_key_value_heads": 2,
+    "pad_token_id": null,
+    "rms_norm_eps": 1e-06,
+    "rope_parameters": {
+      "mrope_section": [
+        1,
+        1,
+        2
+      ],
+      "rope_theta": 500000.0,
+      "rope_type": "default",
+      "type": "default"
+    },
+    "use_cache": true,
+    "vocab_size": 151936
+  },
+  "tie_word_embeddings": false,
+  "transformers_version": "5.8.0",
+  "video_token_id": 151656,
+  "vision_config": {
+    "deepstack_visual_indexes": [
+      0
+    ],
+    "depth": 2,
+    "hidden_act": "gelu_pytorch_tanh",
+    "hidden_size": 16,
+    "in_channels": 3,
+    "initializer_range": 0.02,
+    "intermediate_size": 16,
+    "model_type": "qwen3_vl_vision",
+    "num_heads": 2,
+    "num_position_embeddings": 64,
+    "out_hidden_size": 16,
+    "patch_size": 4,
+    "spatial_merge_size": 2,
+    "temporal_patch_size": 2
+  },
+  "vision_end_token_id": 151653,
+  "vision_start_token_id": 151652
+}

text_encoder/generation_config.json

@@ -0,0 +1,7 @@
+{
+  "_from_model_config": true,
+  "output_attentions": false,
+  "output_hidden_states": false,
+  "transformers_version": "5.8.0",
+  "use_cache": true
+}

text_encoder/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:811c6e1e1f06b403716b046ec6b30529185aa156194abc405d7b8b5b5de91ce5
+size 19962376

tokenizer/chat_template.jinja

@@ -0,0 +1,7 @@
+{% set image_count = namespace(value=0) %}{% set video_count = namespace(value=0) %}{% for message in messages %}{% if loop.first and message['role'] != 'system' %}<|im_start|>system
+You are a helpful assistant.<|im_end|>
+{% endif %}<|im_start|>{{ message['role'] }}
+{% if message['content'] is string %}{{ message['content'] }}<|im_end|>
+{% else %}{% for content in message['content'] %}{% if content['type'] == 'image' or 'image' in content or 'image_url' in content %}{% set image_count.value = image_count.value + 1 %}{% if add_vision_id %}Picture {{ image_count.value }}: {% endif %}<|vision_start|><|image_pad|><|vision_end|>{% elif content['type'] == 'video' or 'video' in content %}{% set video_count.value = video_count.value + 1 %}{% if add_vision_id %}Video {{ video_count.value }}: {% endif %}<|vision_start|><|video_pad|><|vision_end|>{% elif 'text' in content %}{{ content['text'] }}{% endif %}{% endfor %}<|im_end|>
+{% endif %}{% endfor %}{% if add_generation_prompt %}<|im_start|>assistant
+{% endif %}

tokenizer/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:312c03cb421e18f41c20c3e35f3302479a3a8f8b1f06779fd900326a0d32c22f
+size 11420534

tokenizer/tokenizer_config.json

@@ -0,0 +1,35 @@
+{
+  "add_prefix_space": false,
+  "backend": "tokenizers",
+  "bos_token": null,
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "errors": "replace",
+  "extra_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "is_local": false,
+  "local_files_only": false,
+  "max_length": null,
+  "model_max_length": 32768,
+  "pad_to_multiple_of": null,
+  "pad_token": "<|endoftext|>",
+  "pad_token_type_id": 0,
+  "padding_side": "left",
+  "processor_class": "Qwen2VLProcessor",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

transformer/config.json

@@ -0,0 +1,26 @@
+{
+  "_class_name": "MiniMaxH3Transformer3DModel",
+  "_diffusers_version": "0.40.0.dev0",
+  "attention_head_dim": 16,
+  "audio_in_channels": 8,
+  "ffn_dim": 32,
+  "final_norm_eps": 1e-05,
+  "freq_dim": 8,
+  "hidden_size": 24,
+  "in_channels": 4,
+  "norm_eps": 1e-05,
+  "num_attention_heads": 2,
+  "num_layers": 2,
+  "num_refiner_layers": 2,
+  "patch_size": [
+    1,
+    2,
+    2
+  ],
+  "qk_norm_eps": 1e-05,
+  "rope_freq_dim": 2,
+  "rope_theta": 10000.0,
+  "text_dim": 16,
+  "time_embed_dim": 16,
+  "time_embed_hidden_dim": 24
+}

transformer/diffusion_pytorch_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5113e5ca6897c6f9dbcaa1ea86af267d9cba52d7834a99278147859d03136697
+size 164664

transformer_ref/config.json

@@ -0,0 +1,26 @@
+{
+  "_class_name": "MiniMaxH3Transformer3DModel",
+  "_diffusers_version": "0.40.0.dev0",
+  "attention_head_dim": 16,
+  "audio_in_channels": 8,
+  "ffn_dim": 32,
+  "final_norm_eps": 1e-05,
+  "freq_dim": 8,
+  "hidden_size": 24,
+  "in_channels": 4,
+  "norm_eps": 1e-05,
+  "num_attention_heads": 2,
+  "num_layers": 2,
+  "num_refiner_layers": 2,
+  "patch_size": [
+    1,
+    2,
+    2
+  ],
+  "qk_norm_eps": 1e-05,
+  "rope_freq_dim": 2,
+  "rope_theta": 10000.0,
+  "text_dim": 16,
+  "time_embed_dim": 16,
+  "time_embed_hidden_dim": 24
+}

transformer_ref/diffusion_pytorch_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5dfea8eadd245fadd47fe0ff4530ff0858f1fbd49bdcbab0dc8361755f49f53c
+size 164664

vae/config.json

@@ -0,0 +1,45 @@
+{
+  "_class_name": "AutoencoderKLMiniMaxH3",
+  "_diffusers_version": "0.40.0.dev0",
+  "block_out_channels": [
+    8,
+    16
+  ],
+  "clip_length": 17,
+  "decoder_attention_head_dim": 8,
+  "decoder_ffn_mult": 2,
+  "decoder_norm_eps": 1e-05,
+  "decoder_num_attention_heads": 2,
+  "decoder_num_layers": 2,
+  "decoder_num_register_tokens": 2,
+  "decoder_rope_dim_ratio": 0.75,
+  "decoder_rope_theta": 100.0,
+  "in_channels": 3,
+  "latent_channels": 4,
+  "latents_mean": [
+    0.0,
+    0.0,
+    0.0,
+    0.0
+  ],
+  "latents_std": [
+    1.0,
+    1.0,
+    1.0,
+    1.0
+  ],
+  "layers_per_block": 1,
+  "norm_eps": 1e-06,
+  "norm_num_groups": 8,
+  "out_channels": 3,
+  "spatial_downsample_factors": [
+    4,
+    4
+  ],
+  "spatial_padding_mode": "reflect",
+  "temporal_downsample_factors": [
+    2,
+    2
+  ],
+  "token_drop": 3
+}

vae/diffusion_pytorch_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:74ac6d576cad46acc41199785295c8047b1f6f831194bee7a9b490ce266f0726
+size 346960