Spaces:
Runtime error
Runtime error
Commit History
-
Moves images to the top of readme.
5060043 -
Update README for Nemotron text model
15eac0f -
Add routed Modal multi-model deployment
efee902 -
fix: invalidate stale game assets after deploy
453d293 -
fix: slow player movement
9ccdf94 -
fix: preserve character colors in canvas renderer
cdb1892 -
feat: add layered animated player sprite
ba3d37e -
feat: redesign world map
1d6a6bd
-
docs: clarify judge inference setup
f8b86f7 -
fix: install git for Space builds
1ce293c -
fix: pin Space Docker base image
49c16a4 -
docs: finalize submission links
0d8f21e -
docs: prepare Build Small submission
e7f936b -
fix: prioritize typed answers for model grading
42e70ad -
fix: polish answer feedback and tablet controls
8f8996c -
fix: fall back when preloaded models are missing
2f6b4a0 -
chore: make model preload configurable
9f71e26 -
feat: add dual-model inference runtime
3f9808f -
test: tolerate quoted auth cookies
ec35f84 -
Revert "fix: harden generated lesson quality"
This reverts commit f325ce50bbea0bb760101b976592ab4ba8cf3ba0.3cf23f7 -
fix: harden generated lesson quality
f325ce5 -
feat: polish Secret Student game experience
7aeedfb -
fix: preserve authentication in Space iframe
Co-authored-by: OpenAI Codex <codex@openai.com>0e014b2 -
chore: add bulk Space environment configurator
Co-authored-by: OpenAI Codex <codex@openai.com>f1ca79c -
merge: add Space and Modal LLM deployment modes
Merge the Codex-assisted embedded llama.cpp and external Modal runtime support.f4785f1 -
feat: add embedded llama.cpp and Modal LLM runtimes
Select mock, external, or embedded llama.cpp inference at startup; bundle the pinned server-cuda-b9049 image for GPU Spaces, add a Modal deployment, local CUDA Compose support, health reporting, tests, and deployment documentation. Implemented collaboratively with OpenAI Codex. Co-authored-by: OpenAI Codex <codex@openai.com>f0013ff -
chore: store PNG assets with Git LFS and Xet
Track PNG assets through Git LFS so GitHub can retain them and Hugging Face can transfer them through Xet without changing the working image bytes. Implemented collaboratively with OpenAI Codex. Co-authored-by: OpenAI Codex <codex@openai.com>6a60ce0 -
chore: configure Hugging Face Space deployment
Preserve the Space metadata and LFS rules while switching deployment to the repository Dockerfile on port 7860. Implemented collaboratively with OpenAI Codex. Co-authored-by: OpenAI Codex <codex@openai.com>541d9fd -
merge: Codex-assisted Secret Student improvements
Merge the multi-screen game experience and structured LLM tracing work from codex/multi-screen-game-ui.8ee3287 -
feat: polish Secret Student game flow and LLM reliability
Build the multi-screen classroom, home, HQ, and world experience; add resilient structured model output, trace capture, JSON repair, and schema-constrained vision verification. Implemented collaboratively with OpenAI Codex. Co-authored-by: OpenAI Codex <codex@openai.com>f17d2b7
-
initial commit
d0f4f8b