codearena-rl / server

Commit History

Add Ollama fallback support to ai_fixer
05d943b

havinashpatil commited on

Add Hugging Face Inference API fallback for AI fixer
18261cc

havinashpatil commited on

Fix builtin syntax fixer for inline one-line headers and validate TGI fixes
0c0a8ff

havinashpatil commited on

Fix TGI endpoint plumbing and use full TGI runtime image in Dockerfile
a8bc575

havinashpatil commited on

Add TGI integration for cloud LLM serving
434afdf

havinashpatil commited on

Serve React frontend from FastAPI backend for one-click access
62c7e95

havinashpatil commited on

Finalizing CodeArena RL Benchmark: frontend improvements, GRPO training scripts, and cleaned environment
03a7eb9

havinashpatil commited on

Complete all tasks: Adaptive curriculum, GRPO, React frontend, LLM-as-a-judge
a448db8

havinashpatil commited on

Upload meta
03defc2

havinashpatil commited on

fix: reset task_id parsing, grader tuple crash fallback, and inference score output
646409d

adityanaikhpt commited on

fix: clamp scores to strict (0.001, 0.999) range
b3485c6

adityanaikhpt commited on

fix: openenv.yaml tasks graders + strict score bounds
9967cb5

adityanaikhpt commited on

fix: add safe_reward() clamp at every reward return point
f14f8d9

adityanaikhpt commited on

Add normalize_reward(); enforce strict (0,1) on all paths
b53855f

adityanaikhpt commited on

Update grading logic to strict (0,1) ratio fallback
78b7e6b

adityanaikhpt commited on

fix: OpenEnv multi-mode compliance — add main() entrypoint and uv.lock
e92bfc1

adityanaikhpt commited on

Production-ready: add server/app.py with fallback-safe /reset, fix Dockerfile, add HF metadata, add task JSON files
dcc8fa3

adityanaikhpt commited on