Claude Sonnet 5.5 36.66

Anthropic · Claude Code · 1M

Score
36.66
Rank
2/ 10
Best on
4/ 8 tasks
API cost
$406
Input tokens · 99.4% cached
2494 M
Output tokens
9.8 M
TaskJudgedRankTierSelf-bestRoundsHoursAPI cost
LLaVA2.80×2 / 10N12.74×1810.0 h$39.7trace →
YOCO2.26×2 / 10N1—1810.0 h$43.8trace →
NanoChat1.21×1 / 10N11.19×2410.0 h$48.4trace →
BAGEL SFT2.21×1 / 10N12.24×1410.0 h$46.7trace →
R31.96×2 / 10N01.92×910.0 h$37.9trace →
3DGS4.44×2 / 10N14.06×2010.0 h$56.5trace →
Slime GRPO2.40×1 / 10N12.45×1010.0 h$58.3trace →
Slime OPD1.76×1 / 10N11.68×610.0 h$75.2trace →
↑ Leaderboard