Reasoning
198-question Diamond subset; independent AA evaluation
- 01
- 02
- 03
- 03
- 03
- 15
Claude Opus 4.7
Across the task benchmarks we track, Claude Opus 4.7’s strongest relative results are in reasoning, knowledge work, and work automations. On GPQA Diamond, it scores 91.4%, ranking #15 among 30 models with reported results in our dataset.
Updated: Data as of:
Strongest results
| Benchmark | Measures | Score |
|---|---|---|
| MCP-Atlas ↗April 2026 / public 500 | Tool use | 80.8%#2 of 5 |
| Humanity's Last Exam ↗Stable HLE — text-only subset | Expert reasoning | 42.3%Joint #26 of 68 |
| GPQA ↗Diamond / Artificial Analysis | Reasoning | 91.41%#15 of 30 |
| AA-Briefcase ↗1.1 / combined Elo | Business deliverables | 1,252 Elo#33 of 57 |
Model details
| Model | Claude Opus 4.7 |
|---|---|
| Agent Index v1.2 | Insufficient coverage (1/7) · Preview |
| Creator | Anthropic (opens in a new tab) |
| Availability | Proprietary |
| Context size | 1,000,000 tokens (opens in a new tab) |
| Max output | 128,000 tokens (opens in a new tab) |
| Knowledge cutoff | 2026-01-01 |
| License | Not reported |
Pricing and performance
Per provider and reported configuration. Prices are USD per 1M tokens.
| Provider / configuration | Contexttokens | InputUSD / 1M tokens | OutputUSD / 1M tokens | LatencyTTFT · seconds | Speedtokens / sec |
|---|---|---|---|---|---|
| Claude Opus 4.7 (max) | 1,000,000 | $5.00 (opens in a new tab) | $25.00 (opens in a new tab) | 14.738 (opens in a new tab) | 68.472 (opens in a new tab) |
| Claude Opus 4.7 (max) | 1,000,000 | $5.00 (opens in a new tab) | $25.00 (opens in a new tab) | 33.055 (opens in a new tab) | 53.476 (opens in a new tab) |
| Claude Opus 4.7 (max) | 1,000,000 | $5.00 (opens in a new tab) | $25.00 (opens in a new tab) | 33.588 (opens in a new tab) | 53.379 (opens in a new tab) |
| Claude Opus 4.7 (max) | 1,000,000 | $5.00 (opens in a new tab) | $25.00 (opens in a new tab) | 23.76 (opens in a new tab) | 68.4 (opens in a new tab) |
Benchmark performance
Claude Opus 4.7 in color, alongside leading models. Latest stored reports, best effort per report date.
GPQA Diamond ↗
Work Automations
Private held-out set; API mode; 50 steps; strict pass/fail
- 01
- 02
- 03
- 04
- 05
- 15
Knowledge Work
220 tasks; Stirrup agent; pairwise Elo
- 01
- 02
- 03
- 04
- 05
- 26
Reported benchmarks
All published results in our dataset, with versions, evaluation settings, and sources.
| Benchmark | Score | Evaluation setup | Source |
|---|---|---|---|
| AA-Briefcase1.1 / combined EloCombined Elo · Higher is better | 1,252 Elo | selection: best reported configuration per release · Claude Opus 4.7 (max) · tools: benchmark-defined | Artificial Analysis (opens in a new tab)independent |
| AppWorldTest-ChallengeTask Goal Completion · Higher is better | 85.6% | agent: LARA MAS · metric: TGC · selection: All 23 leaderboard submissions; latest report date and best score at that date shown per model. Normal-split and SGC values retained in the source data/settings, not mixed into the ranking. · source model label: claude-opus-4-7 · split: Test-Challenge · test challenge sgc: 77 · test normal sgc: 82.1 · test normal tgc: 88.7 | AppWorld research team ↗ (opens in a new tab)benchmark publisher · 2026-09-02 |
| AutomationBench1.0.6 / Zapier held-out private setStrict task completion · Higher is better | 5.48% | Claude Opus 4.7 (Low); private held-out 657 tasks; API mode; 50-step cap; strict completion | Zapier ↗ (opens in a new tab)benchmark publisher |
| AutomationBench1.0.6 / Zapier held-out private setStrict task completion · Higher is better | 6.85% | Claude Opus 4.7 (Medium); private held-out 657 tasks; API mode; 50-step cap; strict completion | Zapier ↗ (opens in a new tab)benchmark publisher |
| AutomationBench1.0.6 / Zapier held-out private setStrict task completion · Higher is better | 9.13% | Claude Opus 4.7 (High); private held-out 657 tasks; API mode; 50-step cap; strict completion | Zapier ↗ (opens in a new tab)benchmark publisher |
| AutomationBench1.0.6 / Zapier held-out private setStrict task completion · Higher is better | 9.74% | Claude Opus 4.7 (XHigh); private held-out 657 tasks; API mode; 50-step cap; strict completion | Zapier ↗ (opens in a new tab)benchmark publisher |
| AutomationBench1.0.6 / Zapier held-out private setStrict task completion · Higher is better | 13.39% | Claude Opus 4.7 (Max); private held-out 657 tasks; API mode; 50-step cap; strict completion | Zapier ↗ (opens in a new tab)benchmark publisher |
| GDPval-AA2.1Elo · Higher is better | 1,361.69 Elo | Claude Opus 4.7 (Max); 220 tasks; Stirrup agent; pairwise Elo | Artificial Analysis (opens in a new tab)independent |
| GPQADiamond / Artificial AnalysisAccuracy · Higher is better | 91.414% | Claude Opus 4.7 (Max); 198-question Diamond subset; independent AA evaluation | Artificial Analysis (opens in a new tab)independent |
| GPQADiamond / Artificial AnalysisAccuracy · Higher is better | 88.485% | Claude Opus 4.7 (Non-reasoning, High); 198-question Diamond subset; independent AA evaluation | Artificial Analysis (opens in a new tab)independent |
| Humanity's Last ExamStable HLE — full setAccuracy · Higher is better | 54.7% | Launch comparison; GPT models evaluated at xhigh effort; research environment. Competitor settings follow source footnotes. · subset: full · Tools enabled | OpenAI ↗ (opens in a new tab)provider reported · 2026-04-23 |
| Humanity's Last ExamStable HLE — full setAccuracy · Higher is better | 46.9% | Launch comparison; GPT models evaluated at xhigh effort; research environment. Competitor settings follow source footnotes. · subset: full · No tools | OpenAI ↗ (opens in a new tab)provider reported · 2026-04-23 |
| Humanity's Last ExamStable HLE — text-only subsetAccuracy · Higher is better | 42.3% | question count: 2158 · reported precision decimals: 1 · Claude Opus 4.7 (max); original HLE; 2,158 text-only questions; no tools; pass@1; chart labels rounded to one decimal. · source model href: /models/claude-opus-4-7 · subset: text · No tools | Artificial Analysis (opens in a new tab)independent |
| MCP-AtlasApril 2026 / public 500Task success · Higher is better | 80.8% | max; public 500; 100 tool calls | Scale AI ↗ (opens in a new tab)benchmark publisher |