ThinkWAM checkpoints
Project page · Code · Evaluation instructions
ThinkWAM inference checkpoints exported from the selected training checkpoints.
Each .pt contains the model weights and its embedded configuration; optimizer,
scheduler and RNG states are excluded. The original tensor values, storage precision and embedded
configuration are preserved.
| Benchmark | Checkpoint |
|---|---|
| LIBERO / LIBERO-Plus | libero_thinkwam_checkpoint.pt |
| RoboTwin 2.0 | robotwin2.0_thinkwam_checkpoint.pt |
| RoboCasa24 | robocasa24_thinkwam_checkpoint.pt |
Loading and evaluation
Use the ThinkWAM evaluation commands with --checkpoint-profile released,
following the release guide.
The compatible loader checks the pinned artifact/config identity and strict-loads
all policy parameters, excluding only LIBERO's eight inactive legacy phase-head
parameters whose memory contribution is configured to zero. The two other
checkpoints have no phase head.
These published checkpoints predate the current paper training recipes, and
some of their training and sampling settings differ from those recipes; the
release evaluation path preserves this metadata and records
paper_recipe_match: false. Loading compatibility is not an independent
reproduction of the paper's success rates. Full benchmark evaluations still
require the simulators, original normalization statistics, assets and complete
case schedules. See the code repository for validation coverage and limitations.
The V-JEPA encoder is embedded in the policy. Supply a local Qwen2.5-0.5B model
and tokenizer directory with --qwen-path; runtime asset paths are relocatable.
File sizes and SHA256 checksums are recorded in checkpoint_manifest.json and
SHA256SUMS.
License
This checkpoint distribution is released under Apache License 2.0. Embedded Qwen2.5-0.5B tensors retain Apache 2.0 terms, and V-JEPA 2.1 tensors retain their MIT terms and attribution; see NOTICE. The separately released code and third-party code components retain their respective licenses.