SmolDataEnvs train logs
GRPO training logs for Qwen3.5 on SmolDataEnvs
RL Environments at Scale
GRPO training logs for Qwen3.5 on SmolDataEnvs
Extract text, layout, or answer questions from documents
Transcribe multilingual audio and get instant scoring
Visualize RL experiment metrics in an interactive dashboard
Run OpenCode data analysis tasks and get JSON results
Run a coding agent on curated data tasks
Explore tasks, run sandbox tools, and submit answers
Cross-model RL training, evaluation, and run history
Transcribe math images to LaTeX and get a score
From an idea to a trained 4B, with the dead ends left in
Display interactive tracking data visualization
Show real-time I/O activity dashboard
Every painting of every run, with the sketch that made it
Display interactive visualization of tracking data
Interact with a GeoGuessrβstyle environment through custom actions
Show interactive tracking dashboard for your tasks
Interact with a watercolour simulation via text commands
Show an interactive tracking dashboard
Show real-time I/O activity visualization