Search public pages, research tools, and SAVRN solutions.
Open-weight model · Robotics
by Vladislav Sovrasov sovrasov/flux3_policy_black_cubes_5k
flux3_policy_black_cubes_5k is an open-weight model for robotics from Vladislav Sovrasov, released under Apache License 2.0. Its published files total 471.8 MB.
FLUX 3 Action (flux3) is Black Forest Labs' video-action model: the FLUX.3 video trunk with an action modality that is denoised jointly with the next video frames, finetuned per robot embodiment with fresh action heads.
By Vladislav Sovrasov, published under apache-2.0, revision e23bf5898966.
FLUX 3 Action (flux3) is Black Forest Labs' video-action model: the FLUX.3 video trunk with an action modality that is denoised jointly with the next video frames, finetuned per robot embodiment with fresh action heads. This policy has been trained and pushed to the Hub using LeRobot. Learn how to train and run it in the LeRobot flux3 guide, or browse the full documentation. The policy consumes these observation features and produces these action features. Inputs Outputs New to LeRobot? These guides cover the full workflow: - Install LeRobot — set up the lerobot package. - Hardware setup — assemble, wire, and calibrate your robot and cameras. - Record data & train a policy — the end-to-end…
FLUX 3 Action (flux3) is Black Forest Labs' video-action model: the FLUX.3 video trunk with an action modality that is denoised jointly with the next video frames, finetuned per robot embodiment with fresh action heads.
This policy has been trained and pushed to the Hub using LeRobot.
Learn how to train and run it in the LeRobot flux3 guide, or browse the full documentation.
PhysicalAIRobottop, wristThe policy consumes these observation features and produces these action features.
Inputs
| Feature | Type | Shape |
|---|---|---|
observation.images.scene |
VISUAL | (3, 256, 256) |
observation.images.wrist |
VISUAL | (3, 256, 256) |
observation.state |
STATE | (6,) |
Outputs
| Feature | Type | Shape |
|---|---|---|
action |
ACTION | (6,) |
| Setting | Value |
|---|---|
| Training steps | 5000 |
| Batch size | 2 |
| Optimizer | adamw |
| Learning rate | 0.0001 |
| Seed | 42 |
| LeRobot version | 0.6.2 |
New to LeRobot? These guides cover the full workflow:
lerobot package.lerobot-* commands.The short version to run and train this policy:
lerobot-rollout \
--strategy.type=base \
--robot.type=PhysicalAIRobot \
--robot.port=<your_robot_port> \
--robot.cameras="{ <camera_1>: {type: opencv, index_or_path: <index_or_path>, width: 640, height: 480, fps: 30}, <camera_2>: {type: opencv, index_or_path: <index_or_path>, width: 640, height: 480, fps: 30}}" \
--policy.path=sovrasov/flux3_policy_black_cubes_5k \
--task="pick a black cube and move it to the cardboard bin" \
--duration=60
Replace the remaining <...> placeholders with your own values: --robot.port and the camera names/indices are specific to your machine, and the camera names must match the observation keys this policy was trained on.
When --strategy.type=base is used the script doesn't record the episodes. Skipping duration will make the policy run indefinitely. For more information look at rollout documentation.
lerobot-train \
--dataset.repo_id=${HF_USER}/<dataset> \
--policy.type=flux3 \
--output_dir=outputs/train/<policy_repo_id> \
--job_name=lerobot_training \
--policy.device=cuda \
--policy.repo_id=${HF_USER}/<policy_repo_id> \
--wandb.enable=true
Writes checkpoints to outputs/train/<policy_repo_id>/checkpoints/.
No evaluation results have been provided for this policy yet.
If you use this policy, please cite the method linked in the description above, along with LeRobot:
@misc{cadene2024lerobot,
author = {Cadene, Remi and Alibert, Simon and Soare, Alexander and Gallouedec, Quentin and Zouitine, Adil and Palma, Steven and Kooijmans, Pepijn and Aractingi, Michel and Shukor, Mustafa and Aubakirova, Dana and Russi, Martino and Capuano, Francesco and Pascal, Caroline and Choghari, Jade and Moss, Jess and Wolf, Thomas},
title = {LeRobot: State-of-the-art Machine Learning for Real-World Robotics in Pytorch},
howpublished = "\url{https://github.com/huggingface/lerobot}",
year = {2024}
}
11 files, 471.8 MB in total. The weights are 4 files totalling 471.8 MB in safetensors.
| File | Type | Size | SHA-256 |
|---|---|---|---|
| adapter_model.safetensors | Weights | 471.8 MB | aa9c23951e99 |
| policy_postprocessor_step_0_flux3_action_history_unnormalizer.safetensors | Weights | 184 B | 96a5f8de8c33 |
| policy_preprocessor_step_3_flux3_observation_history_normalizer.safetensors | Weights | 360 B | fb0b16f0b790 |
| policy_preprocessor_step_4_flux3_action_target_normalizer.safetensors | Weights | 184 B | 96a5f8de8c33 |
| adapter_config.json | Configuration | 1.4 KB | — |
| config.json | Configuration | 3.5 KB | — |
| policy_postprocessor.json | Configuration | 536 B | — |
| policy_preprocessor.json | Configuration | 1.6 KB | — |
| train_config.json | Configuration | 10.2 KB | — |
| README.md | Documentation | 5.7 KB | — |
| .gitattributes | Repository | 1.5 KB | — |
Released by Vladislav Sovrasov through its official repository on Hugging Face. Read the license.
| Precision | Weights in memory |
|---|---|
| As published | 471.8 MB |
Weights only, from the published parameter count; the key-value cache and runtime add to this.
Yes. flux3_policy_black_cubes_5k is released under Apache License 2.0. The Apache License 2.0 is a permissive open-source license. It permits commercial use, modification and redistribution. It requires keeping the license and copyright notices and any NOTICE file, stating significant changes, and it includes an express patent grant from contributors.
LIBERO 4in1（liberospatial / liberoobject / liberogoal / libero10）共 53.19 GB 的 Wan2.2-VAE 编码 latent 缓存： 训练 LIBERO policy（action head / VLA）时直接读取 latent 缓存，避免重复 VAE 编码。 - 窗口模式: windowed（--windowed） - 输出:.pt 文件，每 episode 一个 - dataset（推荐）: https://huggingface.co/datasets/MangoGoes/libero4in1wan2.2vaelatentdataset - model（本仓库）: https://huggingface.co/MangoGoes/libero4in1wan2.2vaelatentcosmosstyle
Wan2.2-TI2V-5B video DiT + 48-joint action head, trainingmode=joint. The base is suhyeok's finalized B-method recipe: a teacher-forced (sigma=0.25) self-EMA teacher plus an iBOT prototype loss at L18 L18, gamma=0.01, two-view. On top of it the 3 PAST cond latent frames are pooled into one motion frame before a chosen block. These are NOT the surrogate ctxpool runs. The surrogate line (older base, pd8 x GA1) lives in hmkang/wamctxpoolxattn and hmkang/wamctxpoolavg. Do not compare across the two sets. Geometry: 4-latin (numframesin=25, numframesout=41, fdf 2) = 4 cond + 2 future latent slots, 96 tokens per latent frame, 576 tokens per row. Effective batch 16 clips x GA 2 x 2 views = 64 rows…
weighted/imatrix quants of https://huggingface.co/DeepCybo/PhysBrain1.5-8B For a convenient overview and download list, visit our model page for this model. static quants are available at https://huggingface.co/mradermacher/PhysBrain1.5-8B-GGUF This is a vision model - mmproj files (if any) will be in the static repository. If you are unsure how to use GGUF files, refer to one of TheBloke's READMEs for more details, including on how to concatenate multi-part files. (sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants) Here is a handy graph by ikawrakow comparing some lower-quality quant And here are Artefact2's thoughts on the matter…
static quants of https://huggingface.co/MirilAI/Miril-DroneVLM-2B-2 For a convenient overview and download list, visit our model page for this model. weighted/imatrix quants seem not to be available (by me) at this time. If they do not show up a week or so after the static ones, I have probably not planned for them. Feel free to request them by opening a Community Discussion. If you are unsure how to use GGUF files, refer to one of TheBloke's READMEs for more details, including on how to concatenate multi-part files. (sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants) Here is a handy graph by ikawrakow comparing some lower-quality quant…
Own lineage: PPO from scratch (nb1, commit 46c48b1) continued as nb2 and nb4 (commit fac88cc), checkpoint 850 (1.328 s); evolution strategies on the weights against the Arena sprint scorer (0.952 s); then the network frozen and wrapped in a step counter carried in the headroll action that adds joint-target and forward-command offsets from about 0.5 m to the line, found by CMA-ES on the event's official seeds: 0.940 s. No other entry's weights used. Simulation only. A perpetual policy for the microduck (61-D observation, 14 actions, 50 Hz). Runs until told otherwise — a gait, loaded into a policy slot. The observation normalizer is baked into policy.onnx; feed raw observations. manifest.json…