Claude Sonnet 4.5 benchmark ranking on every AI leaderboard

Name: Claude Sonnet 4.5
Author: Anthropic

closed weights · #25 overall, podium score 6.0 · top 10 on 3 of 13 · released Sep 29, 2025

Anthropic

13 of 13 boards

Rank	Board	Category	Score	Of
2	BFCLAgents and tools · as “Claude-Sonnet-4-5-20250929 (FC)”	Agents and tools	73.2%	81
19	MultiChallengeChat · as “claude-sonnet-4-5-20250929-thinking”	Chat	55.3%	29
21	HLEReasoning · as “claude-sonnet-4-5-20250929-thinking”	Reasoning	13.7%	42
23	Document ArenaVision · as “claude-sonnet-4-5-20250929”	Vision	1,450	38
24	Search ArenaAgents and tools · as “claude-sonnet-4-5-search”	Agents and tools	1,158	33
28	MCP AtlasAgents and tools · as “claude-sonnet-4-5 (thinking)”	Agents and tools	59.5%	34
42	ARC-AGI-2Reasoning · as “Claude Sonnet 4.5 (Thinking 32K)”	Reasoning	13.6%	78
47	Text ArenaChat · as “claude-sonnet-4-5-20250929-high-32k”	Chat	1,456	100
63	Epoch ECIReasoning · as “Claude Sonnet 4.5”	Reasoning	146.8	100
67	GPQA DiamondReasoning · as “claude-sonnet-4-5-20250929_59K”	Reasoning	82.3%	100
68	WebDev ArenaCoding · as “claude-sonnet-4-5-20250929-high-32k”	Coding	1,393	100
9	SWE-bench VerifiedCoding · stale since 2026-02-26 · as “Claude 4.5 Sonnet (high)”	Coding	71.4%	39
10	MMLU-ProReasoning · stale since 2026-03-11 · as “Claude-4.5-Sonnet(Thinking)”	Reasoning	87.4%	100

Category ranks and more models

Rank in each category

More from Anthropic

Sources: each leaderboard's own published results page, captured every day at 04:30 UTC. Scores exactly as each board published them; we run no evaluations. Logos via logo.dev; trademarks belong to their owners.

Rank in each category

More from Anthropic

New #1s by email