Model card
Claude Opus 4.5.
AnthropicapiUndisclosed params1 current SOTA
§ 01 · Benchmarks
Every benchmark Claude Opus 4.5 has a recorded score for.
| # | Benchmark | Area · Task | Metric | Value | Rank | Date | Source |
|---|---|---|---|---|---|---|---|
| 01 | SWE-Bench | Computer Code · Code Generation | resolve-rate-agentic | 80.9% | #1 | 2026-02-01 | source ↗ |
| 02 | SWE-Bench Verified | Computer Code · Code Generation | resolve-rate | 80.9% | #2 | 2025-11-24 | source ↗ |
| 03 | SWE-bench Verified | Agentic AI · SWE-bench | resolve-rate | 80.9% | #2 | — | source ↗ |
| 04 | MMLU-Pro | Reasoning · Commonsense Reasoning | accuracy | 89.5% | #3 | 2026-04-20 | source ↗ |
| 05 | SWE-Bench | Computer Code · Code Generation | resolve-rate | 80.9% | #3 | 2026-02-01 | source ↗ |
| 06 | AIME 2025 | Reasoning · Mathematical Reasoning | accuracy | 80.0% | #4 | — | source ↗ |
| 07 | SWE-Bench | Computer Code · Code Generation | resolve-rate-agentic | 79.2% | #5 | 2026-01-01 | source ↗ |
| 08 | MMLU | Reasoning · Commonsense Reasoning | accuracy | 91.6% | #5 | 2025-11-24 | source ↗ |
| 09 | SWE-Bench | Computer Code · Code Generation | resolve-rate | 78.0% | #8 | 2025-12-01 | source ↗ |
| 10 | GSM8K | Reasoning · Mathematical Reasoning | accuracy | 98.6% | #8 | — | source ↗ |
| 11 | SWE-Bench | Computer Code · Code Generation | resolve-rate-agentic | 76.8% | #10 | 2026-02-01 | |
| 12 | GPQA | Reasoning · Multi-step Reasoning | accuracy | 74.9% | #13 | — | source ↗ |
| 13 | MATH | Reasoning · Mathematical Reasoning | accuracy | 90.7% | #15 | — | source ↗ |
Rank column shows this model’s position vs all other models scored on the same benchmark + metric (competitors after the slash). #1 in red means current SOTA. Sorted by rank, then newest result.
§ 02 · Strengths by area
Where Claude Opus 4.5 actually performs.
§ 03 · Papers
1 paper with results for Claude Opus 4.5.
- 2023-10-10· Computer Code· 2 results
SWE-bench: Can Language Models Resolve Real-World GitHub Issues?
Carlos E. Jimenez, John Yang, Alexander Wettig, Shunyu Yao et al.
§ 04 · Related models
Other Anthropic models scored on Codesota.
Claude Opus 4
Undisclosed params · 13 results · 2 SOTA
Claude Opus 4.5
3 results · 2 SOTA
Claude Sonnet 5
Undisclosed params · 2 results · 2 SOTA
Claude Sonnet 4
10 results · 1 SOTA
Claude Mythos Preview
1 result · 1 SOTA
Claude 3.5 Sonnet
Undisclosed params · 27 results
Claude 3.7 Sonnet
10 results
Claude 3 Opus
5 results
§ 05 · Sources & freshness
Where these numbers come from.
anthropic-model-card
5
results
anthropic-internal
1
result
anthropic-blog
1
result
editorial
1
result
pricepertoken
1
result
swebench-leaderboard
1
result
live-swe-agent
1
result
sota-timeline
1
result
mini-swe-agent-v2
1
result
11 of 13 rows marked verified. · first result 2025-11-24, latest 2026-04-20.