test: cover temperature tolerance gate and token×beam cap through translate

fix: cap token×beam product so generation fits its GPU reservation

fix: harden generation params, fix swap RTL, polish from review

docs: sync README test counts (58 fast / 68 total)

fix: greedy-decode by default; apply review cleanups

test: cover bfloat16/float32 dtype selection and requirements invariant

perf: engage ZeroGPU pack/stream, tune GPU duration, fix requirements

test: cover Textbox toolbar buttons and update test counts

chore: upgrade dependencies to latest (hold transformers 4.x)

docs: use uv run in documented commands

docs: update README for 418 languages and dev setup

docs: rename project title to MADLAD-400 Translate

docs: update test counts in CLAUDE.md and README.md

docs: update architecture description and test counts

docs: update README test counts

docs: update architecture description and test counts

docs: update test counts in README

docs: migrate setup instructions from pip/venv to uv

docs: update test counts and gitignore for batch tab redesign

docs: update CLAUDE.md and README.md with new test counts

docs: update CLAUDE.md and README.md for Tier 1 languages

docs: update CLAUDE.md and README.md for CUDA-only support

Update docs for evaluated-only language set

Update Gradio SDK version to 6.8.0 in README

Update README.md

Update README.md

Update README.md

Update README.md

Update README.md

initial commit