🐛 Fix OSS designation for gpt and claude models

✨ Add Opus 4.8+OpenCode

🐛 Change from a simple to a weighted average

🐛 Specify thinking level for gpt 5.5

✨ Add GPT 5.5 + Codex

✨ Add Opus 4.8

🐛 Remove 'Ensure HF Space exists' step from CI

✨ Add openclaw benchmarks

✨ Add Qwen Code runs

🐛 Clean up benchmark runs

🐛 Remove environment; fix slider

🎨 Clean up leaderboard

🐛 Fix slider

🐛 Present scores out of 100

✨ Add claude-sonnet-4-6-claude-code

✨ Add swe-bench-pro--ansible-qwen3-6-35b-nvfp4-opencode

🐛 Make swe-bench-verified the first column

📝 Update docstring

🐛 Add patch to gradio_leaderboard before import

🧪 Test switching leaderboard for plain dataframe

🐛 Switch from gr.Row to gr.Tabs

✨ Add simplified leaderboard and other benchmarks

👷 Add stage github workflow

🐛 Fix score precision to 3 decimal places

✨ Add new/updated metrics

🐛 Remove skills from intro text

🚸 Change columns to title-case

🐛 Remove skills - for now

✨ Add visual formatting and themeing

🚸 Use plain harness names; move harness to col 2

🚸 Make asterisk more pronounced; simplify tabs

🐛 Make links open in new tab; show model_id

🎉 Reset repo