Robotics
PEFT
Safetensors
jevany
lora
decision-model
pointer
reward-model
judge
calibration
conformal-prediction
Instructions to use SimpleJev/JevAny-Robotics-4B-LoRA with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- PEFT
How to use SimpleJev/JevAny-Robotics-4B-LoRA with PEFT:
from peft import PeftModel from transformers import AutoModel base_model = AutoModel.from_pretrained("Qwen/Qwen3.5-4B") model = PeftModel.from_pretrained(base_model, "SimpleJev/JevAny-Robotics-4B-LoRA") - Notebooks
- Google Colab
- Kaggle
Commit History
14 days ago
-
Card: link the training set, SimpleJev/JevAny-Robotics-Train-v0.4
Verified signature62b19e1 -
v0.4: LoRA trained from Qwen3.5-4B on JevAny-Robotics-Train-v0.4 (CC BY-NC-SA 4.0); v0.2 stays available as revision v0.2
Verified signature9ff40ab
27 days ago
-
Model card: same-harness comparison with dedicated robot reward models; --media full usage
Verified signatureAdds RoboReward-8B/4B, Robometer-4B and Robo-Dopamine-GRM-2.0-8B (all run in our harness), calibration, policy ranking, speed and media-encoding results; removes leaderboard comparisons; usage now requires JevAny --media full.0894a55
-
Card: use the JevAny-* repository names (renamed 2026-10-17); data files unchanged
Verified signature41e3cd9 -
Update README.md
Verified signature236d876
-
Update README.md
Verified signaturee25ce9f -
Update README.md
Verified signature5b4e444 -
Update README.md
Verified signature5c4d17e -
model card: full baseline comparison (same harness) + paired CIs; add eval/comparison reports
Verified signaturea52d6ed -
robojudge-4b-v02 SFT final (step 5780)
Verified signaturefac5b68 -
initial commit
Verified signature86bc6bd