Publish real-world checkpoint and runtime assets

README.md

@@ -1,3 +1,71 @@
 ---
 license: mit
+library_name: pytorch
+tags:
+  - robotics
+  - streamingwam
+  - real-world
+tasks:
+  - robotics
 ---
+
+# StreamingWAM real-world checkpoint
+
+This repository contains the StreamingWAM checkpoint used for the seven-task
+Franka real-robot evaluation. It is separate from the
+[LIBERO simulation checkpoint](https://huggingface.co/rsj2003/StreamingWAM-LIBERO)
+and the [real-world demonstration dataset](https://huggingface.co/datasets/rsj2003/StreamingWAM-RealWorld).
+
+## Files
+
+- `streamingwam_realworld.pt`: complete model weights (one file);
+- `streamingwam-realworld-runtime-assets.tar.gz`: normalization statistics and cached embeddings for the seven task prompts;
+- `streamingwam_realworld_model.yaml`: model-only deployment configuration;
+- `checkpoint_info.json` and `SHA256SUMS`: artifact metadata and integrity checks.
+
+The runtime archive contains `train_split_zscore_stats.json` and
+`text_embeds_cache/`. No training logs or optimizer artifacts are added.
+
+The matching implementation and model-only deployment configuration are in
+the public [StreamingWAM repository](https://github.com/renshaojie233/StreamingWAM/tree/main/real_robot).
+
+## Download and use
+
+From the [StreamingWAM code repository](https://github.com/renshaojie233/StreamingWAM):
+
+```bash
+python real_robot/download_assets.py --output checkpoints/streamingwam-realworld
+```
+
+The downloader verifies SHA-256 checksums and installs the checkpoint locally
+as `step_030000.pt` for compatibility with the existing runtime. See the
+[real-robot setup guide](https://github.com/renshaojie233/StreamingWAM/tree/main/real_robot)
+for observation preprocessing, shadow-mode inference, and hardware integration.
+
+## Contract
+
+- control frequency: 15 Hz;
+- cameras: `camera_1`, `camera_3`, `camera_2`;
+- each camera: RGB `224 x 128`, vertically concatenated to `3 x 384 x 224`;
+- image history: five consecutive frames;
+- action prefix: sixteen actually executed native actions;
+- output: sixteen future actions of dimension eight;
+- arm action: `(q[t+1] - q[t]) / 0.2`;
+- gripper action/state: continuous absolute command, `0=closed`, `1=open`;
+- initial warm start: ten denoising steps;
+- rolling update: one denoising substep;
+- supported trained replanning horizons: `K={1,2,4,8,10}`.
+
+The released checkpoint is the standard model, not a latency-shift fine-tune.
+No transition blending or trajectory smoothing is part of the model artifact.
+
+## Integrity
+
+`streamingwam_realworld.pt` is 12,041,739,641 bytes and has SHA-256:
+
+```text
+ed68b39a9d7025188e6216a9336a3e265893d360b0c02b91f31629c317d1f66a
+```
+
+The Wan 2.2 TI2V 5B base model is not redistributed and must be obtained under
+its original license.

SHA256SUMS

@@ -0,0 +1,4 @@
+ed68b39a9d7025188e6216a9336a3e265893d360b0c02b91f31629c317d1f66a  streamingwam_realworld.pt
+251466fc7be4883f2e4c3196f0c9522e52d5f12123a9cc3eed9cc6d5f78007ed  streamingwam-realworld-runtime-assets.tar.gz
+939f27d83c8977cbb4be4367bf0f2b1d32bed0724aac2e84c91693456a17dd3d  streamingwam_realworld_model.yaml
+836dbf8578a8ecaa97784bf93da43e95a9ce42ed372ab8185e3ae6ae81f75ef2  checkpoint_info.json

checkpoint_info.json

@@ -0,0 +1,10 @@
+{
+  "model": "StreamingWAM",
+  "domain": "real-world",
+  "filename": "streamingwam_realworld.pt",
+  "sha256": "ed68b39a9d7025188e6216a9336a3e265893d360b0c02b91f31629c317d1f66a",
+  "size_bytes": 12041739641,
+  "action_dim": 8,
+  "source_release": "realworld-v1",
+  "migration": "Unchanged checkpoint moved from GitHub Releases to Hugging Face."
+}

streamingwam-realworld-runtime-assets.tar.gz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:251466fc7be4883f2e4c3196f0c9522e52d5f12123a9cc3eed9cc6d5f78007ed
+size 5120095

streamingwam_realworld.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ed68b39a9d7025188e6216a9336a3e265893d360b0c02b91f31629c317d1f66a
+size 12041739641

streamingwam_realworld_model.yaml

@@ -0,0 +1,60 @@
+model:
+  _target_: fastwam.runtime.create_fastwam_joint
+  model_id: Wan-AI/Wan2.2-TI2V-5B
+  tokenizer_model_id: Wan-AI/Wan2.1-T2V-1.3B
+  tokenizer_max_len: 128
+  load_text_encoder: false
+  proprio_dim: 8
+  redirect_common_files: true
+  mot_checkpoint_mixed_attn: false
+  action_dit_pretrained_path: null
+  skip_dit_load_from_pretrain: true
+  num_history_frames: 5
+  video_dit_config:
+    has_image_input: false
+    patch_size: [1, 2, 2]
+    in_dim: 48
+    hidden_dim: 3072
+    ffn_dim: 14336
+    freq_dim: 256
+    text_dim: 4096
+    out_dim: 48
+    num_heads: 24
+    attn_head_dim: 128
+    num_layers: 30
+    eps: 1.0e-6
+    seperated_timestep: true
+    require_clip_embedding: false
+    require_vae_embedding: false
+    fuse_vae_embedding_in_latents: true
+    use_gradient_checkpointing: false
+    video_attention_mask_mode: first_frame_causal
+    action_conditioned: false
+    action_dim: 8
+    action_group_causal_mask_mode: group_diagonal
+  action_dit_config:
+    action_dim: 8
+    hidden_dim: 1024
+    ffn_dim: 4096
+    num_heads: 24
+    attn_head_dim: 128
+    num_layers: 30
+    text_dim: 4096
+    freq_dim: 256
+    eps: 1.0e-6
+    use_gradient_checkpointing: false
+  video_scheduler:
+    train_shift: 5.0
+    infer_shift: 5.0
+    num_train_timesteps: 1000
+  action_scheduler:
+    train_shift: 5.0
+    infer_shift: 5.0
+    num_train_timesteps: 1000
+  loss:
+    lambda_action: 1.0
+  action_clean_prefix_steps: 16
+  sdp_future_action_train_prob: 0.8
+  sdp_future_action_steps_choices: [1, 2, 4, 8, 10]
+  sdp_future_num_noise_levels: 32
+  sdp_future_random_offset: true