Spaces:

MaduRox
/

Kalpana-Multi-Model-Benchmark

Paused

App Files Files Community

Kalpana-Multi-Model-Benchmark

Ctrl+K

Ctrl+K

1 contributor

History: 23 commits

MaduRox's picture

Add on-bar token labels + reduction-factor callout (RIF's advantage was invisible at linear scale); add sample-size caveat to accuracy chart

f63f356 verified about 1 month ago

.pytest_cache
Upload folder using huggingface_hub about 1 month ago
__pycache__
Upload folder using huggingface_hub about 1 month ago
benchmark
Show visible progress during synthetic generation; cap token count to selected model's real context window (prevents GPU OOM during RIF pack compilation); replace raw exception dumps with clean, actionable error messages about 1 month ago
clients
Fix RIF chat/completions parsing: API returns SSE stream, not plain JSON, for real (non-mock) queries about 1 month ago
tests
Upload folder using huggingface_hub about 1 month ago
ui
Add on-bar token labels + reduction-factor callout (RIF's advantage was invisible at linear scale); add sample-size caveat to accuracy chart about 1 month ago
.dockerignore

96 Bytes
Upload folder using huggingface_hub about 1 month ago
.gitattributes

1.52 kB
initial commit about 1 month ago
Dockerfile

505 Bytes
Upload folder using huggingface_hub about 1 month ago
README.md

1.2 kB
Upload folder using huggingface_hub about 1 month ago
app.py

560 Bytes
Upload folder using huggingface_hub about 1 month ago
requirements.txt

222 Bytes
Pin fastapi/starlette to versions compatible with gradio 4.44.1's Jinja2Templates usage (starlette 1.x broke template loader hashing) about 1 month ago