Anthropic

As of 2026-10-18, Anthropic has 37 models on the meo leaderboard, 4 first-party-tested. Its highest-ranked model is Anthropic: Claude Opus 5.5 (#3, 92.1%).

Official: www.anthropic.com

meo rank	Model	Accuracy	𝕍 index	$ / correct	Context	Weights
#3	Anthropic: Claude Opus 5.5	92.1%	100.0	$0.0165	1M	—
#6	Anthropic: Claude Fable 5	88.7%	27.5	$0.0573	1M	—
#7	Anthropic: Claude Opus 4.8	84.9%	12.1	$0.0459	1M	—
#77	Anthropic: Claude 3 Haiku	41.8%	0.014	$0.0009	200K	—
—	Anthropic: Claude 3.5 Haiku	—	—	—	200K	—
—	Anthropic: Claude Fable 5 (batch)	—	—	—	1M	—
—	Anthropic: Claude Fable 5.1	—	—	—	1M	—
—	Anthropic: Claude Fable 5.1 (batch)	—	—	—	1M	—
—	Anthropic: Claude Haiku 4.5	—	—	—	200K	—
—	Anthropic: Claude Haiku 4.5 (batch)	—	—	—	200K	—
—	Anthropic: Claude Haiku 5.5	—	—	—	1M	—
—	Anthropic: Claude Haiku 5.5 (batch)	—	—	—	1M	—
—	Anthropic: Claude Opus 4	—	—	—	200K	—
—	Anthropic: Claude Opus 4.1	—	—	—	200K	—
—	Anthropic: Claude Opus 4.1 (batch)	—	—	—	200K	—
—	Anthropic: Claude Opus 4.5	—	—	—	200K	—
—	Anthropic: Claude Opus 4.5 (batch)	—	—	—	200K	—
—	Anthropic: Claude Opus 4.6	—	—	—	1M	—
—	Anthropic: Claude Opus 4.6 (batch)	—	—	—	1M	—
—	Anthropic: Claude Opus 4.6 (Fast)	—	—	—	1M	—
—	Anthropic: Claude Opus 4.7	—	—	—	1M	—
—	Anthropic: Claude Opus 4.7 (batch)	—	—	—	1M	—
—	Anthropic: Claude Opus 4.7 (Fast)	—	—	—	1M	—
—	Anthropic: Claude Opus 4.8 (batch)	—	—	—	1M	—
—	Anthropic: Claude Opus 4.8 (Fast)	—	—	—	1M	—
—	Anthropic: Claude Opus 5	—	—	—	1M	—
—	Anthropic: Claude Opus 5 (batch)	—	—	—	1M	—
—	Anthropic: Claude Opus 5.5 (batch)	—	—	—	1M	—
—	Anthropic: Claude Sonnet 4	—	—	—	200K	—
—	Anthropic: Claude Sonnet 4.5	—	—	—	1M	—
—	Anthropic: Claude Sonnet 4.5 (batch)	—	—	—	1M	—
—	Anthropic: Claude Sonnet 4.6	—	—	—	1M	—
—	Anthropic: Claude Sonnet 4.6 (batch)	—	—	—	1M	—
—	Anthropic: Claude Sonnet 5	—	—	—	1M	—
—	Anthropic: Claude Sonnet 5 (batch)	—	—	—	1M	—
—	Anthropic: Claude Sonnet 5.5	—	—	—	1M	—
—	Anthropic: Claude Sonnet 5.5 (batch)	—	—	—	1M	—

Frequently asked questions

How many AI models does Anthropic have?

Anthropic has 37 models on the meo leaderboard, of which 4 are first-party-tested. Its best-ranked model is Anthropic: Claude Opus 5.5 at #3 (92.1% accuracy).

What is Anthropic's best AI model?

On the un-leaked meo benchmark, Anthropic's top model is Anthropic: Claude Opus 5.5, ranked #3 overall with 92.1% accuracy.

Does Anthropic release open-weights models?

None of Anthropic's listed models are flagged open-weights; they are served via API.

Related: all AI labs · full leaderboard · rankings · methodology.