Knowledge · Benchmark profile

Artificial Analysis MMLU-Pro

An independently evaluated MMLU-Pro result from Artificial Analysis.

Reference2026Active3 models

Data verified 1 Aug 2026 · Methodology 2.3.0

Benchmark score on Artificial Analysis MMLU-Pro

Leaderboard · 3 models

Claude Opus 4.5 Thinking leads at 89.5%, followed by Claude Opus 4.5 (88.9%) and GPT-OSS 120B (80.8%).

Display onlyKnowledge2026

#1Closed

Claude Opus 4.5 Thinking

Anthropic

claude-opus-4-5-thinking

89.5%

Overall — · Context 200K tokens

#2Closed

Claude Opus 4.5

Anthropic

claude-opus-4-5

88.9%

Overall #57 · Context 200K tokens

#3Open

GPT-OSS 120B

OpenAI

gpt-oss-120b

80.8%

Overall #84 · Context 131K tokens

ViewModelsColours

AnthropicOpenAI

Lumina

Artificial Analysis MMLU-Pro leaderboard

Accessible data for Artificial Analysis MMLU-Pro leaderboard
Model	Rank	Model ID	Provider	Score	Unit
Claude Opus 4.5 Thinking	#1	claude-opus-4-5-thinking	Anthropic	89.5	percent
Claude Opus 4.5	#2	claude-opus-4-5	Anthropic	88.9	percent
GPT-OSS 120B	#3	gpt-oss-120b	OpenAI	80.8	percent

Benchmark score table (3 models)

One best compatible score per canonical product · higher is better

Artificial Analysis MMLU-Pro leaderboard ranked by score
Rank	Model	Provider	License	Evidence use	Score
#1	Claude Opus 4.5 Thinking claude-opus-4-5-thinking	Anthropic	closed	Source carrier	89.5%
#2	Claude Opus 4.5 claude-opus-4-5	Anthropic	closed	Source carrier	88.9%
#3	GPT-OSS 120B gpt-oss-120b	OpenAI	open	Source carrier	80.8%

The top of this snapshot is led by Claude Opus 4.5 Thinking at 89.5%; third place is 8.7 points behind. The top-3 spread is 8.7 points.

Configuration-preserving evidencePublished source rows (9)Exact source labels, configurations, versions and dated observations. Same-snapshot canonical/configuration projections are shown once.

Every distinct dated source-published row for Artificial Analysis MMLU-Pro, including exact model configuration labels
Exact published model / configuration	Canonical product	Score	Version / system	Evidence	Source / observed
Exact BenchLM registry variant Claude Opus 4.5 Thinking; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Claude Opus 4.5 Thinking	89.5%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Claude Opus 4.5; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Claude Opus 4.5	88.9%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant GPT-OSS 120B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-OSS 120B	80.8%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Claude Opus 4.5 Thinking; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Claude Opus 4.5 Thinking	89.5%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Claude Opus 4.5; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Claude Opus 4.5	88.9%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant GPT-OSS 120B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-OSS 120B	80.8%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Claude Opus 4.5 Thinking; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Claude Opus 4.5 Thinking	89.5%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21
Exact BenchLM registry variant Claude Opus 4.5; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Claude Opus 4.5	88.9%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21
Exact BenchLM registry variant GPT-OSS 120B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-OSS 120B	80.8%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Compare top models

Claude Opus 4.5 Thinking vs Claude Opus 4.5 Claude Opus 4.5 vs GPT-OSS 120B

About Artificial Analysis MMLU-Pro

Definition and scoring

Organisation: Artificial Analysis
Category: Knowledge
Version: 2026
Direction: higher is better
Ranking use: Reference
Contamination risk: Unknown

An independently evaluated MMLU-Pro result from Artificial Analysis. Every genuine source row stays tied to its exact model label, configuration, benchmark version and evaluation system. The summary chart shows one best compatible score per canonical product; Score 2.0 admits only explicitly mapped, frozen protocols and keeps incompatible configurations separate.

Open benchmark source ↗

FAQ

Quick answers

What does Artificial Analysis MMLU-Pro measure?

An independently evaluated MMLU-Pro result from Artificial Analysis.

Which model scores highest on Artificial Analysis MMLU-Pro?

Claude Opus 4.5 Thinking by Anthropic currently leads with 89.5%.

How many models are evaluated on Artificial Analysis MMLU-Pro?

3 canonical model products have a qualifying score, backed by 9 distinct dated source rows.

Knowledge · Benchmark profile

Artificial Analysis MMLU-Pro

An independently evaluated MMLU-Pro result from Artificial Analysis.

Reference2026Active3 models

Data verified 1 Aug 2026 · Methodology 2.3.0

Benchmark score on Artificial Analysis MMLU-Pro

Leaderboard · 3 models

Claude Opus 4.5 Thinking leads at 89.5%, followed by Claude Opus 4.5 (88.9%) and GPT-OSS 120B (80.8%).

Display onlyKnowledge2026

#1Closed

Claude Opus 4.5 Thinking

Anthropic

claude-opus-4-5-thinking

89.5%

Overall — · Context 200K tokens

#2Closed

Claude Opus 4.5

Anthropic

claude-opus-4-5

88.9%

Overall #57 · Context 200K tokens

#3Open

GPT-OSS 120B

OpenAI

gpt-oss-120b

80.8%

Overall #84 · Context 131K tokens

ViewModelsColours

AnthropicOpenAI

Lumina

Artificial Analysis MMLU-Pro leaderboard

Accessible data for Artificial Analysis MMLU-Pro leaderboard
Model	Rank	Model ID	Provider	Score	Unit
Claude Opus 4.5 Thinking	#1	claude-opus-4-5-thinking	Anthropic	89.5	percent
Claude Opus 4.5	#2	claude-opus-4-5	Anthropic	88.9	percent
GPT-OSS 120B	#3	gpt-oss-120b	OpenAI	80.8	percent

Benchmark score table (3 models)

One best compatible score per canonical product · higher is better

Artificial Analysis MMLU-Pro leaderboard ranked by score
Rank	Model	Provider	License	Evidence use	Score
#1	Claude Opus 4.5 Thinking claude-opus-4-5-thinking	Anthropic	closed	Source carrier	89.5%
#2	Claude Opus 4.5 claude-opus-4-5	Anthropic	closed	Source carrier	88.9%
#3	GPT-OSS 120B gpt-oss-120b	OpenAI	open	Source carrier	80.8%

The top of this snapshot is led by Claude Opus 4.5 Thinking at 89.5%; third place is 8.7 points behind. The top-3 spread is 8.7 points.

Configuration-preserving evidencePublished source rows (9)Exact source labels, configurations, versions and dated observations. Same-snapshot canonical/configuration projections are shown once.

Every distinct dated source-published row for Artificial Analysis MMLU-Pro, including exact model configuration labels
Exact published model / configuration	Canonical product	Score	Version / system	Evidence	Source / observed
Exact BenchLM registry variant Claude Opus 4.5 Thinking; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Claude Opus 4.5 Thinking	89.5%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Claude Opus 4.5; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Claude Opus 4.5	88.9%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant GPT-OSS 120B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-OSS 120B	80.8%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Claude Opus 4.5 Thinking; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Claude Opus 4.5 Thinking	89.5%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Claude Opus 4.5; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Claude Opus 4.5	88.9%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant GPT-OSS 120B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-OSS 120B	80.8%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Claude Opus 4.5 Thinking; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Claude Opus 4.5 Thinking	89.5%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21
Exact BenchLM registry variant Claude Opus 4.5; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Claude Opus 4.5	88.9%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21
Exact BenchLM registry variant GPT-OSS 120B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-OSS 120B	80.8%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Compare top models

Claude Opus 4.5 Thinking vs Claude Opus 4.5 Claude Opus 4.5 vs GPT-OSS 120B

About Artificial Analysis MMLU-Pro

Definition and scoring

Organisation: Artificial Analysis
Category: Knowledge
Version: 2026
Direction: higher is better
Ranking use: Reference
Contamination risk: Unknown

Open benchmark source ↗

FAQ

Quick answers

What does Artificial Analysis MMLU-Pro measure?

An independently evaluated MMLU-Pro result from Artificial Analysis.

Which model scores highest on Artificial Analysis MMLU-Pro?

Claude Opus 4.5 Thinking by Anthropic currently leads with 89.5%.

How many models are evaluated on Artificial Analysis MMLU-Pro?

3 canonical model products have a qualifying score, backed by 9 distinct dated source rows.

Navigation

Artificial Analysis MMLU-Pro

Leaderboard · 3 models

Claude Opus 4.5 Thinking

Claude Opus 4.5

GPT-OSS 120B

Artificial Analysis MMLU-Pro leaderboard

Crop Artificial Analysis MMLU-Pro leaderboard

Benchmark score table (3 models)

Crop Artificial Analysis MMLU-Pro benchmark table

Crop Artificial Analysis MMLU-Pro published source rows

Compare top models

Definition and scoring

Quick answers

What does Artificial Analysis MMLU-Pro measure?

Which model scores highest on Artificial Analysis MMLU-Pro?

How many models are evaluated on Artificial Analysis MMLU-Pro?

Artificial Analysis MMLU-Pro

Leaderboard · 3 models

Claude Opus 4.5 Thinking

Claude Opus 4.5

GPT-OSS 120B

Artificial Analysis MMLU-Pro leaderboard

Crop Artificial Analysis MMLU-Pro leaderboard

Benchmark score table (3 models)

Crop Artificial Analysis MMLU-Pro benchmark table

Crop Artificial Analysis MMLU-Pro published source rows

Compare top models

Definition and scoring

Quick answers

What does Artificial Analysis MMLU-Pro measure?

Which model scores highest on Artificial Analysis MMLU-Pro?

How many models are evaluated on Artificial Analysis MMLU-Pro?

Leaderboard · 3 models

Claude Opus 4.5 Thinking

Claude Opus 4.5

GPT-OSS 120B

Benchmark score table (3 models)

Compare top models

Definition and scoring

Quick answers

What does Artificial Analysis MMLU-Pro measure?

Which model scores highest on Artificial Analysis MMLU-Pro?

How many models are evaluated on Artificial Analysis MMLU-Pro?

More Knowledge benchmarks

Leaderboard · 3 models

Claude Opus 4.5 Thinking

Claude Opus 4.5

GPT-OSS 120B

Benchmark score table (3 models)

Compare top models

Definition and scoring

Quick answers

What does Artificial Analysis MMLU-Pro measure?

Which model scores highest on Artificial Analysis MMLU-Pro?

How many models are evaluated on Artificial Analysis MMLU-Pro?

More Knowledge benchmarks