Claude Opus 4.7 vs Claude Opus 4.8
Different efforts: high · max · Compare all at high
Humanity's Last Exam Humanity's Last Exam: Very hard expert questions across many fields, written to stump today’s models
Opus 4.733.3
Opus 4.848.7
Arena Text · overall Arena Text · overall: Human pairwise preferences in the stated text vote pool
Opus 4.71,501
Opus 4.8 · high1,481
Arena Text · Chinese Arena Text · chinese: Human pairwise preferences in the stated text vote pool
Opus 4.71,531.8
Opus 4.81,506.8
Arena Text · English Arena Text · english: Human pairwise preferences in the stated text vote pool
Opus 4.71,496.3
Opus 4.8 · high1,468.6
Arena Text · French Arena Text · french: Human pairwise preferences in the stated text vote pool
Opus 4.71,503
Opus 4.8 · high1,481.1
Arena Text · German Arena Text · german: Human pairwise preferences in the stated text vote pool
Opus 4.71,495.2
Opus 4.8 · high1,472.4
Arena Text · instruction following Arena Text · instruction following: Human pairwise preferences in the stated text vote pool
Opus 4.71,498
Opus 4.8 · high1,476.3
Arena Text · Japanese Arena Text · japanese: Human pairwise preferences in the stated text vote pool
Opus 4.71,473.9
Opus 4.8 · high1,441.1
Arena Text · Korean Arena Text · korean: Human pairwise preferences in the stated text vote pool
Opus 4.71,450.9
Opus 4.8 · high1,431.9
Arena Text · non-English Arena Text · non english: Human pairwise preferences in the stated text vote pool
Opus 4.71,480.5
Opus 4.8 · high1,450
Arena Text · Polish Arena Text · polish: Human pairwise preferences in the stated text vote pool
Opus 4.71,479.5
Opus 4.81,459.3
Arena Text · Russian Arena Text · russian: Human pairwise preferences in the stated text vote pool
Opus 4.71,494.4
Opus 4.8 · high1,474
Arena Text · Spanish Arena Text · spanish: Human pairwise preferences in the stated text vote pool
Opus 4.71,494.6
Opus 4.8 · high1,465
AA Omniscience Index AA Omniscience Index: Scores factual knowledge while penalizing incorrect answers to encourage admitting uncertainty
Opus 4.714.8
Opus 4.828.8
SimpleQA Verified v1 SimpleQA Verified v1: Accuracy on short factual questions with verified answers, answered without web access
Opus 4.7 · xhigh51.7
Opus 4.853
CritPt (physics) CritPt (physics): Research-level physics problems that need many careful reasoning steps
Opus 4.75.1
Opus 4.820.9
AA Intelligence Index 4.3.2 AA Intelligence Index 4.3.2: Artificial Analysis combines tests of knowledge, reasoning, coding and tool use into one score
Opus 4.730.9
Opus 4.841.8
AA Omniscience · Accuracy AA Omniscience · Accuracy: Measures correct answers on the AA Omniscience factual knowledge test
Opus 4.744.7
Opus 4.848.8
AA Omniscience · Hallucination Rate AA Omniscience · Hallucination Rate: Measures how often attempted answers are incorrect on the AA Omniscience factual knowledge test · lower is better
Opus 4.754.1
Opus 4.839.3
AA-LCR (long context) 1.1 AA-LCR (long context) 1.1: Questions that need reading and connecting facts across very long documents (around 100K tokens)
Opus 4.775.7
Opus 4.877.7
IFBench (instructions) IFBench (instructions): Checks whether the model follows precise formatting and content instructions
Opus 4.743.6
Opus 4.862.2
MLCR-AA MLCR-AA: Models synthesize long medical records; the headline credits concise answers judged both accurate and complete
Opus 4.728.9
Opus 4.845.6
EQ-Bench 3 EQ-Bench 3: Judge-rated responses to social and emotionally nuanced scenarios, scored through the source pairwise ranking
Opus 4.71,884.3
Opus 4.82,029.8
Chess Puzzles Chess Puzzles: Models choose the best next move from a novel chess position
Opus 4.7 · max41
Opus 4.852
FrontierMath FrontierMath: Models solve expert-written advanced mathematics problems on the named tier and edition
Opus 4.7 · xhigh43.8
Opus 4.847.2
FrontierMath Tier 4 FrontierMath Tier 4: Models solve expert-written advanced mathematics problems on the named tier and edition
Opus 4.7 · xhigh22.9
Opus 4.831.3
FrontierMath Tier 4 (v2) FrontierMath Tier 4 (v2): Models solve expert-written advanced mathematics problems on the named tier and edition
Opus 4.7 · max31.7
Opus 4.856.1
FrontierMath Tiers 1–3 (v2) FrontierMath Tiers 1–3 (v2): Models solve expert-written advanced mathematics problems on the named tier and edition
Opus 4.7 · max70.2
Opus 4.880
Mystery Game Puzzles Mystery Game Puzzles: Models choose the best move in puzzle positions from an undisclosed game
Opus 4.7 · max28
Opus 4.836
MathArena · Apex MathArena · apex · Apex: Answer accuracy on the named mathematics competition under the recorded effort
Opus 4.7 · xhigh40.6
Opus 4.881.3
MathArena · Apex Shortlist MathArena · apex · Apex Shortlist: Answer accuracy on the named mathematics competition under the recorded effort
Opus 4.7 · xhigh63.8
Opus 4.890.4
MathArena · ArXivMath · April 2026 MathArena · arxiv · 04/2026: Answer accuracy on the named mathematics competition under the recorded effort
Opus 4.7 · xhigh57.5
Opus 4.862.5
MathArena · ArXivMath · February 2026 MathArena · arxiv · 02/2026: Answer accuracy on the named mathematics competition under the recorded effort
Opus 4.7 · xhigh40.6
Opus 4.860.2
MathArena · ArXivMath · March 2026 MathArena · arxiv · 03/2026: Answer accuracy on the named mathematics competition under the recorded effort
Opus 4.7 · xhigh50.8
Opus 4.875
MathArena · BrokenArXiv · April 2026 MathArena · arxiv_false · 04/2026: Answer accuracy on the named mathematics competition under the recorded effort
Opus 4.7 · xhigh4.1
Opus 4.834.4
MathArena · BrokenArXiv · February 2026 MathArena · arxiv_false · 02/2026: Answer accuracy on the named mathematics competition under the recorded effort
Opus 4.7 · xhigh4
Opus 4.835.5
MathArena · BrokenArXiv · March 2026 MathArena · arxiv_false · 03/2026: Answer accuracy on the named mathematics competition under the recorded effort
Opus 4.7 · xhigh5.8
Opus 4.835.7
Vals LegalBench Vals LegalBench: Legal reasoning and legal-language understanding on LegalBench
Opus 4.785.3
Opus 4.883.6
Vals MedCode Vals MedCode: Medical coding accuracy on patient encounters
Opus 4.754.9
Opus 4.853.2
Vals MedScribe Vals MedScribe: Accuracy of clinical notes produced from consultation transcripts
Opus 4.783
Opus 4.885.8
Vals Legal Research Bench Vals Legal Research Bench: Grounded legal research; a task passes only if all criteria pass
Opus 4.738.5
Opus 4.843.8
GDP.pdf · AA all-pass GDP.pdf · AA all-pass: Models reason over professional PDF documents; an attempt passes only when every atomic criterion passes
Opus 4.7—
Opus 4.822.8
ARC-AGI-1 · semi-private ARC-AGI-1 · semi-private: Models infer transformations from examples and solve unseen grid puzzles; the evaluation split is explicit
Opus 4.7—
Opus 4.892.5
ARC-AGI-2 · semi-private ARC-AGI-2 · semi-private: Models infer transformations from examples and solve unseen grid puzzles; the evaluation split is explicit
Opus 4.7—
Opus 4.8 · high72.1
ARC-AGI-3 · semi-private ARC-AGI-3 · semi-private: Measures interactive game solving through relative human action efficiency
Opus 4.7—
Opus 4.8 · high1.5
MathArena · ArXivMath · January 2026 MathArena · arxiv · 01/2026: Answer accuracy on the named mathematics competition under the recorded effort
Opus 4.7 · xhigh52.2
Opus 4.8—
MathArena · ArXivMath · June 2026 MathArena · arxiv · 06/2026: Answer accuracy on the named mathematics competition under the recorded effort
Opus 4.7—
Opus 4.870
MathArena · ArXivMath · May 2026 MathArena · arxiv · 05/2026: Answer accuracy on the named mathematics competition under the recorded effort
Opus 4.7—
Opus 4.865
MathArena · BrokenArXiv · June 2026 MathArena · arxiv_false · 06/2026: Answer accuracy on the named mathematics competition under the recorded effort
Opus 4.7—
Opus 4.841.7
MathArena · BrokenArXiv · May 2026 MathArena · arxiv_false · 05/2026: Answer accuracy on the named mathematics competition under the recorded effort
Opus 4.7—
Opus 4.823
MathArena · ArXivLean · June 2026 MathArena · arxivlean · 06/2026: Answer accuracy on the named mathematics competition under the recorded effort
Opus 4.7—
Opus 4.823.9
MathArena · ArXivLean · March 2026 MathArena · arxivlean · 03/2026: Answer accuracy on the named mathematics competition under the recorded effort
Opus 4.714.6
Opus 4.8—
MathArena · Kangaroo 2025 1-2 MathArena · kangaroo · Kangaroo 2025 1-2: Answer accuracy on the named mathematics competition under the recorded effort
Opus 4.7—
Opus 4.870.8
MathArena · Kangaroo 2025 11-12 MathArena · kangaroo · Kangaroo 2025 11-12: Answer accuracy on the named mathematics competition under the recorded effort
Opus 4.7—
Opus 4.8100
MathArena · Kangaroo 2025 3-4 MathArena · kangaroo · Kangaroo 2025 3-4: Answer accuracy on the named mathematics competition under the recorded effort
Opus 4.7—
Opus 4.860.4
MathArena · Kangaroo 2025 5-6 MathArena · kangaroo · Kangaroo 2025 5-6: Answer accuracy on the named mathematics competition under the recorded effort
Opus 4.7—
Opus 4.870
MathArena · Kangaroo 2025 7-8 MathArena · kangaroo · Kangaroo 2025 7-8: Answer accuracy on the named mathematics competition under the recorded effort
Opus 4.7—
Opus 4.896.7
MathArena · Kangaroo 2025 9-10 MathArena · kangaroo · Kangaroo 2025 9-10: Answer accuracy on the named mathematics competition under the recorded effort
Opus 4.7—
Opus 4.891.7
| Benchmark | Opus 4.7 | Opus 4.8 |
|---|---|---|
| Humanity's Last Exam | 33.3 | 48.7 |
| Arena Text · overall | 1,501 | 1,481 · high |
| Arena Text · Chinese | 1,531.8 | 1,506.8 |
| Arena Text · English | 1,496.3 | 1,468.6 · high |
| Arena Text · French | 1,503 | 1,481.1 · high |
| Arena Text · German | 1,495.2 | 1,472.4 · high |
| Arena Text · instruction following | 1,498 | 1,476.3 · high |
| Arena Text · Japanese | 1,473.9 | 1,441.1 · high |
| Arena Text · Korean | 1,450.9 | 1,431.9 · high |
| Arena Text · non-English | 1,480.5 | 1,450 · high |
| Arena Text · Polish | 1,479.5 | 1,459.3 |
| Arena Text · Russian | 1,494.4 | 1,474 · high |
| Arena Text · Spanish | 1,494.6 | 1,465 · high |
| AA Omniscience Index | 14.8 | 28.8 |
| SimpleQA Verified v1 | 51.7 · xhigh | 53 |
| CritPt (physics) | 5.1 | 20.9 |
| AA Intelligence Index 4.3.2 | 30.9 | 41.8 |
| AA Omniscience · Accuracy | 44.7 | 48.8 |
| AA Omniscience · Hallucination Rate | 54.1 | 39.3 |
| AA-LCR (long context) 1.1 | 75.7 | 77.7 |
| IFBench (instructions) | 43.6 | 62.2 |
| MLCR-AA | 28.9 | 45.6 |
| EQ-Bench 3 | 1,884.3 | 2,029.8 |
| Chess Puzzles | 41 · max | 52 |
| FrontierMath | 43.8 · xhigh | 47.2 |
| FrontierMath Tier 4 | 22.9 · xhigh | 31.3 |
| FrontierMath Tier 4 (v2) | 31.7 · max | 56.1 |
| FrontierMath Tiers 1–3 (v2) | 70.2 · max | 80 |
| Mystery Game Puzzles | 28 · max | 36 |
| MathArena · Apex | 40.6 · xhigh | 81.3 |
| MathArena · Apex Shortlist | 63.8 · xhigh | 90.4 |
| MathArena · ArXivMath · April 2026 | 57.5 · xhigh | 62.5 |
| MathArena · ArXivMath · February 2026 | 40.6 · xhigh | 60.2 |
| MathArena · ArXivMath · March 2026 | 50.8 · xhigh | 75 |
| MathArena · BrokenArXiv · April 2026 | 4.1 · xhigh | 34.4 |
| MathArena · BrokenArXiv · February 2026 | 4 · xhigh | 35.5 |
| MathArena · BrokenArXiv · March 2026 | 5.8 · xhigh | 35.7 |
| Vals LegalBench | 85.3 | 83.6 |
| Vals MedCode | 54.9 | 53.2 |
| Vals MedScribe | 83 | 85.8 |
| Vals Legal Research Bench | 38.5 | 43.8 |
| GDP.pdf · AA all-pass | — | 22.8 |
| ARC-AGI-1 · semi-private | — | 92.5 |
| ARC-AGI-2 · semi-private | — | 72.1 · high |
| ARC-AGI-3 · semi-private | — | 1.5 · high |
| MathArena · ArXivMath · January 2026 | 52.2 · xhigh | — |
| MathArena · ArXivMath · June 2026 | — | 70 |
| MathArena · ArXivMath · May 2026 | — | 65 |
| MathArena · BrokenArXiv · June 2026 | — | 41.7 |
| MathArena · BrokenArXiv · May 2026 | — | 23 |
| MathArena · ArXivLean · June 2026 | — | 23.9 |
| MathArena · ArXivLean · March 2026 | 14.6 | — |
| MathArena · Kangaroo 2025 1-2 | — | 70.8 |
| MathArena · Kangaroo 2025 11-12 | — | 100 |
| MathArena · Kangaroo 2025 3-4 | — | 60.4 |
| MathArena · Kangaroo 2025 5-6 | — | 70 |
| MathArena · Kangaroo 2025 7-8 | — | 96.7 |
| MathArena · Kangaroo 2025 9-10 | — | 91.7 |
Specs
| Spec | Opus 4.7 | Opus 4.8 |
|---|---|---|
| Input, per 1M | $5 | $5 |
| Output, per 1M | $25 | $25 |
| Cost per task? | — | $4.08 |
| Speed, tok/s | 59 | 66 |
| Context | 1M | 1M |
| Weights | Closed | Closed |
| Released | Apr 2026 | May 2026 |
| See images | ✓ | ✓ |
| Read PDFs | Partial | Partial |
| Hear audio | — | — |
| Watch video | — | — |
| Call tools | ✓ | ✓ |
| Force a tool call | Partial | Partial |
| Return strict JSON | ✓ | ✓ |
| Search the web | ? | ? |
| Run code | ? | ? |
| Use a computer | ? | ? |
What breaks when you switch
Opus 4.7 → Opus 4.8
Nothing known.
Opus 4.8 → Opus 4.7
Nothing known.