Robotics
Safetensors
LeRobot
cosmos3
surgical-robotics
dvrk
surgical-kinematics
motion-prediction
inverse-dynamics
forward-dynamics
action-conditioned
world-model
Instructions to use introvoyz041/cosmos3-nanoh-surgical-dvrk with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- LeRobot
How to use introvoyz041/cosmos3-nanoh-surgical-dvrk with LeRobot:
- Notebooks
- Google Colab
- Kaggle
| { | |
| "architectures": [ | |
| "Cosmos3OmniModel" | |
| ], | |
| "auto_map": { | |
| "AutoConfig": "cosmos3.model.Cosmos3OmniConfig", | |
| "AutoModel": "cosmos3.model.Cosmos3OmniModel" | |
| }, | |
| "dtype": "bfloat16", | |
| "model": { | |
| "_recursive_": false, | |
| "_target": "omni_mot_model", | |
| "config": { | |
| "action_gen": true, | |
| "activation_checkpointing": { | |
| "determinism_check": "default", | |
| "mode": "full", | |
| "preserve_rng_state": true, | |
| "save_ops_regex": [ | |
| "fmha" | |
| ] | |
| }, | |
| "causal_training_strategy": "none", | |
| "compile": { | |
| "compile_dynamic": true, | |
| "compiled_region": "language", | |
| "coordinate_descent_tuning": false, | |
| "enabled": false, | |
| "max_autotune_pointwise": false, | |
| "use_cuda_graphs": false | |
| }, | |
| "diffusion_expert_config": { | |
| "base_fps": 24, | |
| "enable_fps_modulation": true, | |
| "load_weights_from_pretrained": true, | |
| "max_vae_latent_side_after_patchify": 20, | |
| "patch_spatial": 2, | |
| "position_embedding_type": "unified_3d_mrope", | |
| "rope_h_extrapolation_ratio": 1.0, | |
| "rope_t_extrapolation_ratio": 1.0, | |
| "rope_w_extrapolation_ratio": 1.0, | |
| "timestep_range": 1.0, | |
| "unified_3d_mrope_reset_spatial_ids": true, | |
| "unified_3d_mrope_temporal_modality_margin": 15000 | |
| }, | |
| "ema": { | |
| "enabled": false, | |
| "iteration_shift": 0, | |
| "rate": 0.1 | |
| }, | |
| "fixed_step_sampler_config": null, | |
| "input_caption_key": "ai_caption", | |
| "input_image_key": "images", | |
| "input_video_key": "video", | |
| "joint_attn_implementation": "flex", | |
| "latent_downsample_factor": 16, | |
| "lbl": { | |
| "coeff_gen": null, | |
| "coeff_und": null, | |
| "method": "local" | |
| }, | |
| "log_enc_time_every_n": 100, | |
| "lora_alpha": 32, | |
| "lora_enabled": false, | |
| "lora_rank": 16, | |
| "lora_target_modules": "q_proj_moe_gen,k_proj_moe_gen,v_proj_moe_gen,o_proj_moe_gen", | |
| "max_action_dim": 64, | |
| "max_num_tokens_after_packing": 32768, | |
| "natten_parameter_list": null, | |
| "net": null, | |
| "num_embodiment_domains": 32, | |
| "parallelism": { | |
| "cfg_parallel_shard_degree": 1, | |
| "context_parallel_shard_degree": 1, | |
| "data_parallel_replicate_degree": 1, | |
| "data_parallel_shard_degree": -1, | |
| "enable_inference_mode": false, | |
| "fsdp_master_dtype": "float32" | |
| }, | |
| "precision": "bfloat16", | |
| "rectified_flow_inference_config": { | |
| "num_train_timesteps": 1000, | |
| "scheduler_type": "unipc", | |
| "shift": 1, | |
| "use_dynamic_shifting": false | |
| }, | |
| "rectified_flow_training_config": { | |
| "action_loss_weight": 10.0, | |
| "high_sigma_ratio": 0.05, | |
| "high_sigma_timesteps_max": 1000, | |
| "high_sigma_timesteps_min": 995, | |
| "image_loss_scale": 1.0, | |
| "independent_action_schedule": false, | |
| "independent_sound_schedule": false, | |
| "loss_scale": 1.0, | |
| "normalize_loss_by_active": false, | |
| "shift": { | |
| "256": 3, | |
| "480": 5, | |
| "720": 10 | |
| }, | |
| "shift_action": null, | |
| "shift_sound": null, | |
| "sound_loss_scale": null, | |
| "train_time_action_distribution": "logitnormal", | |
| "train_time_image_distribution": "logitnormal", | |
| "train_time_sound_distribution": "logitnormal", | |
| "train_time_video_distribution": "waver", | |
| "train_time_weight": "uniform", | |
| "use_discrete_rf": false, | |
| "use_dynamic_shift": false, | |
| "use_high_sigma_strategy": false, | |
| "use_high_sigma_strategy_action": false, | |
| "use_high_sigma_strategy_sound": false | |
| }, | |
| "resolution": "256", | |
| "sound_dim": null, | |
| "sound_gen": false, | |
| "sound_latent_fps": 25, | |
| "sound_tokenizer": null, | |
| "state_ch": 48, | |
| "state_t": 300, | |
| "tokenizer": { | |
| "_target": "wan2pt2_vae_interface", | |
| "bucket_name": "", | |
| "chunk_duration": 93, | |
| "encode_bucket_multiple": null, | |
| "encode_chunk_frames": { | |
| "256": 68, | |
| "480": 24, | |
| "720": 12 | |
| }, | |
| "encode_exact_durations": null, | |
| "keep_decoder_cache": false, | |
| "object_store_credential_path_pretrained": "", | |
| "spatial_compression_factor": 16, | |
| "temporal_compression_factor": 4, | |
| "temporal_window": null, | |
| "use_streaming_encode": false, | |
| "vae_path": "/workspace/cosmos/checkpoints/wan22_vae/Wan2.2_VAE.pth" | |
| }, | |
| "video_temporal_causal": false, | |
| "vision_gen": true, | |
| "vlm_config": { | |
| "layer_module": "Qwen2MoTDecoderLayer", | |
| "model_instance": { | |
| "_target": "qwen3_vl_text_for_causal_lm", | |
| "config": { | |
| "_target": "create_vlm_config", | |
| "base_config": { | |
| "_target": "qwen3_vl_mot_config_from_json_file", | |
| "json_file": "cosmos3://vfm/models/vlm/qwen3_vl/configs/Qwen3-VL-8B-Instruct.json" | |
| }, | |
| "freeze_und": false, | |
| "layer_module": "MoTDecoderLayer", | |
| "qk_norm_for_text": true, | |
| "tie_word_embeddings": true | |
| } | |
| }, | |
| "model_name": "Qwen/Qwen3-VL-8B-Instruct", | |
| "pretrained_weights": { | |
| "backbone_path": "s3://bucket0/cosmos3/pretrained/huggingface/Qwen/Qwen3-VL-8B-Instruct/", | |
| "checkpoint_format": null, | |
| "credentials_path": "", | |
| "enable_gcs_patch_in_boto3": true, | |
| "enabled": false | |
| }, | |
| "qk_norm": false, | |
| "safetensors_path": "", | |
| "tie_word_embeddings": false, | |
| "tokenizer": { | |
| "_target": "create_qwen2_tokenizer_with_download", | |
| "config_variant": "hf", | |
| "pretrained_model_name": "Qwen/Qwen3-VL-8B-Instruct" | |
| }, | |
| "use_system_prompt": false | |
| } | |
| } | |
| }, | |
| "model_type": "cosmos3_omni", | |
| "transformers_version": "4.57.6" | |
| } |