Python

LiveBench · 28 models

Rank	Model	Score
1	claude-opus-4-8-xhigh-effort	60
1	gpt-5.2-codex	60
3	claude-fable-5-xhigh-effort	55
3	claude-sonnet-5-xhigh-effort	55
3	gemini-3.1-pro-preview-high	55
3	gpt-5.2-2025-12-11-high	55
3	gpt-5.4-mini-xhigh	55
3	gpt-5.5-xhigh	55
3	kimi-k2.6-thinking	55
10	claude-opus-4-6-thinking-auto-high-effort	50
10	claude-sonnet-4-6-thinking-auto-medium-effort	50
10	deepseek-v4-pro	50
10	gemini-3.5-flash-high	50
10	glm-5.2	50
10	gpt-5.4-nano-xhigh	50
10	gpt-5.4-xhigh	50
10	gpt-5.5	50
10	gpt-5.5-high	50
10	kimi-k2.7-code	50
20	claude-opus-4-7-xhigh-effort	45
20	grok-build-0.1	45
20	qwen3.6-plus	45
20	qwen3.7-max	45
24	claude-opus-4-5-20251101-thinking-64k-high-effort	40
24	qwen3.6-27b	40
26	deepseek-v4-flash	35
26	minimax-m3	35
28	grok-4.3	15