claude-sonnet-4-20250514: benchmark scores, price and ranking
claude-sonnet-4-20250514 from Anthropic ranks #306 of 743 LLMs on the LLMs Tiger aggregate benchmark score (48.9), based on 9 benchmark results captured up to 2026-10-13.
- Organisation
- Anthropic
- Type
- Frontier model
- Open weights
- No
- Released
- 2025-05-22 (source: curated)
- Aggregate rank
- #306 of 743 (score 48.9)
Relative to other models, claude-sonnet-4-20250514 is strongest on TheAgentCompany (#2 of 13) and weakest on Fiction.liveBench (#37 of 47).
Across the 8 benchmarks where at least ten models are scored, claude-sonnet-4-20250514 is in the top 10% on 0 and in the top half on 5.
claude-sonnet-4-20250514 benchmark results
| Benchmark | Score | Rank | Source | Captured |
|---|---|---|---|---|
| TheAgentCompany | 33.1% | #2 of 13 | Epoch AI Benchmarking Hub | 2026-10-13 |
| Aider Polyglot | 61.3% | #11 of 49 | Aider polyglot leaderboard | 2026-10-13 |
| Cybench | 35.0% | #5 of 16 | Epoch AI Benchmarking Hub | 2026-10-13 |
| WeirdML | 43.9% | #43 of 109 | Epoch AI Benchmarking Hub | 2026-10-13 |
| OSWorld | 43.9% | #2 of 5 | Epoch AI Benchmarking Hub | 2026-10-13 |
| GSO-Bench | 4.9% | #9 of 19 | Epoch AI Benchmarking Hub | 2026-10-13 |
| GPQA Diamond | 66.7% | #152 of 275 | Epoch AI Benchmarking Hub | 2026-10-13 |
| OTIS Mock AIME | 28.9% | #123 of 189 | Epoch AI Benchmarking Hub | 2026-10-13 |
| Fiction.liveBench | 46.9% | #37 of 47 | Epoch AI Benchmarking Hub | 2026-10-13 |
About claude-sonnet-4-20250514
claude-sonnet-4-20250514 is a frontier model from Anthropic. It was released on 2025-05-22, according to the developer's announcement. No API price is listed for it yet. We hold 9 benchmark results for it across 4 areas, and none is in the top tenth of its board.
Where claude-sonnet-4-20250514 stands
On our aggregate score, claude-sonnet-4-20250514 is #306 of 743 ranked models, around the middle of the table. Its score is 48.9; the scale is centred on 50, so it sits 1.1 points below the typical model. That beats 59% of the ranked field. Among the 70 ranked Anthropic models it is #48. The next one up is Claude Opus 4.1 (unknown thinking) at 49.1. The next one down is claude-opus-4-5-20251101-high-32k at 48.9.
claude-sonnet-4-20250514, benchmark by benchmark
TheAgentCompany puts claude-sonnet-4-20250514 at 33.1%, #2 of 13. That beats 85% of the models tested, but 8.0 points short of gemini-2.5-flash-preview-09-2025 (41.1%). The results either side are gemini-2.5-flash-preview-09-2025 (41.1%) and claude-3-7-sonnet-20250219 (30.9%). It covers work tasks in a simulated software company. The board median is 8.6%, so claude-sonnet-4-20250514 is 24.5 points above it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
On Aider Polyglot, claude-sonnet-4-20250514 reaches 61.3%. It ranks #11 of 49, ahead of 78% of the field, and 21.8 points behind gemini-2.5-pro-preview-06-05 at 83.1%. The results either side are o1-2024-12-17 (61.7%) and Qwen3 235B A22B diff, no think, Alibaba API (59.6%). It covers edits across 6 languages (225 exercises). The board median is 44.9%, so claude-sonnet-4-20250514 is 16.4 points above it. Source: Aider polyglot leaderboard, captured 2026-10-13.
claude-sonnet-4-20250514 manages 35.0% on Cybench, good for #5 of 16. grok-4-0709 leads at 43.0%, 8.0 points further up, and 69% of models sit below. The results either side are Claude Opus 4 (38.0%) and grok-4-fast (30.0%). It covers cybersecurity capture-the-flag tasks. The board median is 17.5%, so claude-sonnet-4-20250514 is 17.5 points above it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
WeirdML puts claude-sonnet-4-20250514 at 43.9%, #43 of 109. That beats 61% of the models tested, but 49.4 points short of GPT-6 Astra (max) (93.3%). The results either side are claude-haiku-4-5-20251001_16K (44.1%) and Claude Opus 4 (43.7%). It covers unusual machine-learning tasks. The board median is 41.2%, so claude-sonnet-4-20250514 is 2.7 points above it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
On GSO-Bench, claude-sonnet-4-20250514 reaches 4.9%. It ranks #9 of 19, ahead of 53% of the field, and 35.3 points behind GPT-5.5 (xhigh) at 40.2%. The results either side are Claude Sonnet 4 (unknown thinking) (4.9%) and Kimi K2 Instruct (4.9%). It covers software performance optimisation. The median score on this board is also 4.9%. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
claude-sonnet-4-20250514 lands below the middle on GPQA Diamond: 66.7%, #152 of 275, with 55% of the field scoring higher. GPT-6 Astra (max) leads at 95.8%; the closest result above is mimo-v2.5-pro at 66.7%. It covers expert-level science questions. The board median is 71.7%, so claude-sonnet-4-20250514 is 5.1 points below it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
claude-sonnet-4-20250514 lands below the middle on OTIS Mock AIME: 28.9%, #123 of 189, with 65% of the field scoring higher. Claude Fable 5.1 (max) leads at 100.0%; the closest result above is magistral-small-2506 at 30.0%. It covers competition maths (AIME 2024-25 style). The board median is 57.8%, so claude-sonnet-4-20250514 is 28.9 points below it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
claude-sonnet-4-20250514 lands below the middle on Fiction.liveBench: 46.9%, #37 of 47, with 77% of the field scoring higher. grok-4-0709 leads at 94.4%; the closest result above is gemini-2.5-flash-lite-preview-06-17 (Default thinking length) at 47.2%. It covers long-context story comprehension (16k). The board median is 61.1%, so claude-sonnet-4-20250514 is 14.2 points below it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
The full table above also lists the other 1 result, including boards with fewer than ten models, where a rank says little.
Strengths and weaknesses
Software engineering: solid, ahead of 65% of the field on average across 2 benchmarks. Its best board here is Aider Polyglot (#11 of 49) and its weakest is GSO-Bench (#9 of 19). Reasoning, maths and science: uneven, ahead of 40% of the field on average across 2 benchmarks. Its best board here is GPQA Diamond (#152 of 275) and its weakest is OTIS Mock AIME (#123 of 189). General knowledge and chat: solid, ahead of 50% of the field on average across 3 benchmarks. Its best board here is Cybench (#5 of 16) and its weakest is Fiction.liveBench (#37 of 47).
Release history
The Anthropic model released before it was claude-opus-4-20250514 on 2025-05-22, which scores 52.5. The next one was claude-sonnet-4-20250514_32K on 2025-05-22, at 49.9.
Reading these numbers
The 9 scores on this page come from 2 sources: Epoch AI Benchmarking Hub (8), Aider polyglot leaderboard (1). 1 of them come from an official leaderboard and 8 from datasets or third-party runs.
Models ranked nearby
- #303 grok-4.20-beta1 (49.0)
- #304 qwen3.7-max-preview (49.0)
- #305 internlm2_5-7b-chat (48.9)
- #307 Qwen3.8 Flash (xhigh) (48.9)
- #308 Frogmini 14B 2510 (48.9)
- #309 Nova Pro (48.9)
More from Anthropic
- Claude Opus 5.5 (max) (#1, 64.3)
- Claude Fable 5.1 (max) (#3, 61.9)
- Claude Sonnet 5.5 (max) (#4, 60.8)
- Claude Opus 5 (max) (#8, 60.0)
- Claude Fable 5 (max) (#9, 59.9)
- claude-opus-4-8-xhigh-effort (#14, 56.6)
Compare it on the live leaderboard → · How the aggregate score works