GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
Amazon-FAR
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Model Details
Model Provider
Amazon-FAR
Model Tree
Qwen/Qwen3-VL-4B-Instruct
Input Modalities
Output Modalities
Supported Functionality
Initialized from Qwen/Qwen3-VL-4B-Instruct at revision ebb281ec70b05090aa6165b016eac8ec08e71b17, as recorded in the StopWatch training recipe. Weights use bf16 safetensors in the stock Qwen3VLForConditionalGeneration layout.
python
import torchfrom transformers import AutoModelForImageTextToText, AutoProcessor repo_id = "Amazon-FAR/stopwatch-perception-4b"model = AutoModelForImageTextToText.from_pretrained( repo_id, dtype=torch.bfloat16)processor = AutoProcessor.from_pretrained(repo_id)The full pipeline code is available at amazon-far/stopwatch. Use the StopWatch code for prompt construction and Habitat-Lab evaluation. Perception performs full prefill of the latest 128 frames every step.
These are paired perception-and-policy results from the paper evaluation archive on the released clean open-vocabulary benchmarks. Our criterion requires STOP with geodesic distance below 0.25 m to any instance viewpoint of an accepted goal class under our vocabulary, measured on the Stretch protocol navmesh. Clean subsets exclude annotation failures and empty or unreachable destination sets.
| Benchmark | Successes / episodes | Success rate |
|---|---|---|
| OVON val_unseen clean | 1606 / 2309 | 69.6% |
| MP3D-OV | 595 / 1193 | 49.9% |
| HSSD-OV | 988 / 1751 | 56.4% |
OVON, MP3D and HSSD frontier-exploration trajectories. See Amazon-FAR/stopwatch-data.
Apache 2.0.
Paper: StopWatch: Goal Grounding for Object-goal Navigation.
bibtex
@article{liu2026stopwatch, title={StopWatch: Goal Grounding for Object-goal Navigation}, author={Liu, Xinhao and Tian, Yunsheng and Qiu, Yuheng and Abbeel, Pieter and Chen, Xi and Shi, Guanya and Feng, Chen}, journal={arXiv preprint arXiv:2610.23841}, year={2026}}@article{liu2026stopwatch, title={StopWatch: Goal Grounding for Object-goal Navigation}, author={Liu, Xinhao and Tian, Yunsheng and Qiu, Yuheng and Abbeel, Pieter and Chen, Xi and Shi, Guanya and Feng, Chen}, journal={arXiv preprint arXiv:2610.23841}, year={2026}}