KAI

Complete Qwen3-VL-2B VLM for LAW + OpenWAM: Q1 planning, Q2 completion/failure assessment, and Q3 action-conditioned value estimation. Includes the frozen backbone, Q1/Q2 adapters, query embeddings and tokenizer at the repository root, plus the matching value checkpoint under q3/. Use this VLM with LAW-OpenWAM. Matching training, inference, resource download and verification code: hey-cjj/LAW. Files are copied from the previously verified checkpoints without changing tensor bytes. Q2 and Q3 are the selected step-50,000 checkpoints. This packaging change does not retrain the model. The release previously passed six bounded 32-GPU two-optimizer-step training save/load checks and two bounded…

World-action model for LAW + OpenWAM, selected step-120,815 checkpoint. Use with the complete LAW-VLM-OpenWAM Q1/Q2/Q3 VLM. Includes WAM weights, configuration and normalization statistics under wam/. The corresponding VLM is packaged separately. Original combined releases remain available in this repository's Git history. Code and download guide: hey-cjj/LAW. Historical RoboTwin overall success: 95.56%. These are selected historical results, not a new benchmark. Six bounded 32-GPU training save/load checks and both bounded real RoboTwin policy episodes passed before this packaging change; see release audit for scope and historical provenance limits. Tensor bytes are unchanged. LAW-authored…

Complete Qwen3-VL-2B VLM for LAW + FastWAM: Q1 planning, Q2 completion/failure assessment, and Q3 action-conditioned value estimation. Includes the frozen backbone, Q1/Q2 adapters, query embeddings and tokenizer at the repository root, plus the matching value checkpoint under q3/. Use this VLM with LAW-FastWAM. Matching training, inference, resource download and verification code: hey-cjj/LAW. Files are copied from the previously verified checkpoints without changing tensor bytes. Q2 and Q3 are the selected step-50,000 checkpoints. This packaging change does not retrain the model. The release previously passed six bounded 32-GPU two-optimizer-step training save/load checks and two bounded…

World-action model for LAW + FastWAM, selected step-30,000 checkpoint. Use with the complete LAW-VLM-FastWAM Q1/Q2/Q3 VLM. Includes WAM weights, configuration and normalization statistics under wam/. The corresponding VLM is packaged separately. Original combined releases remain available in this repository's Git history. Code and download guide: hey-cjj/LAW. Historical RoboTwin overall success: 94.21%. These are selected historical results, not a new benchmark. Six bounded 32-GPU training save/load checks and both bounded real RoboTwin policy episodes passed before this packaging change; see release audit for scope and historical provenance limits. Tensor bytes are unchanged. LAW-authored…

This repository contains uncompressed tar shards for the fastwam dataset. The archive members keep their paths relative to the dataset root. Create a directory named fastwam and extract every file under that directory so the original relative paths are restored, for example: The public inventory is available as DATASETINVENTORY.json.gz, and the current shard state is in SHARDS.json. The inventory contains relative file paths and sizes only; it does not contain the source host path. This upload excludes latent caches and other excluded cache, output, and log directories (9 excluded directories are recorded in the inventory). It also excludes any other files omitted by the inventory.…