Claude Sonnet 5 (max): benchmark scores, price and ranking

Claude Sonnet 5 (max) from Anthropic ranks #55 of 743 LLMs on the LLMs Tiger aggregate benchmark score (52.7), based on 14 benchmark results captured up to 2026-10-13.

Organisation

Anthropic

Type

Frontier model

Open weights

No

Released

2026-06-30 (source: epoch)

Cheapest API price

$2.00 in / $10.00 out per million tokens (via aihubmix, observed 2026-10-13T19:47Z)

Aggregate rank

#55 of 743 (score 52.7)

Relative to other models, Claude Sonnet 5 (max) is strongest on DTBench (#15 of 146) and weakest on CursorBench (#14 of 15).

Across the 14 benchmarks where at least ten models are scored, Claude Sonnet 5 (max) is in the top 10% on 0 and in the top half on 10.

Its cheapest listed input price, $2.00 per million tokens, is lower than that of 18% of the 427 models with a listed price.

Claude Sonnet 5 (max) benchmark results

About Claude Sonnet 5 (max)

Claude Sonnet 5 (max) is a frontier model from Anthropic. It was released on 2026-06-30, according to Epoch AI. The cheapest API listing we track charges $2.00 per million input tokens and $10.00 per million output tokens, through aihubmix. We hold 14 benchmark results for it across 4 areas, and none is in the top tenth of its board.

Where Claude Sonnet 5 (max) stands

On our aggregate score, Claude Sonnet 5 (max) is #55 of 743 ranked models, in the top quarter. Its score is 52.7; the scale is centred on 50, so it sits 2.7 points above the typical model. That beats 93% of the ranked field. Among the 70 ranked Anthropic models it is #14. The next one up is claude-3.5-sonnet-20240620 at 52.7. The next one down is Claude 4.6 Opus at 52.6.

Claude Sonnet 5 (max), benchmark by benchmark

DTBench puts Claude Sonnet 5 (max) at 92.5%, #15 of 146. That beats 90% of the models tested, but 6.5 points short of Claude Opus 5.5 (max) (98.9%). The results either side are GPT-5.6 Terra (max) (93.3%) and Qwen3.7 Max (92.3%). It covers dynamic tool-use benchmark. The board median is 68.8%, so Claude Sonnet 5 (max) is 23.7 points above it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.

On LMCA, Claude Sonnet 5 (max) reaches 49.3%. It ranks #14 of 108, ahead of 87% of the field, and 18.9 points behind Claude Opus 5.5 (max) at 68.2%. The results either side are Grok 4.7 (xhigh) (49.4%) and GPT-5.6 Luna (max) (48.5%). It covers language-model capability assessment. The board median is 29.2%, so Claude Sonnet 5 (max) is 20.1 points above it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.

Claude Sonnet 5 (max) manages 77.0% on ProofBench, good for #7 of 46. Claude Fable 5.1 (max) leads at 100.0%, 23.0 points further up, and 85% of models sit below. The results either side are GPT-5.6 Sol (max) (83.0%) and Tencent Hy4 preview (unknown) (75.0%). It covers writing mathematical proofs. The board median is 21.0%, so Claude Sonnet 5 (max) is 56.0 points above it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.

CritPt puts Claude Sonnet 5 (max) at 16.9%, #26 of 115. That beats 77% of the models tested, but 15.4 points short of GPT-5.6 Sol (max) (32.3%). The results either side are Muse Spark 1.2 (xhigh) (17.7%) and DeepSeek V4 Flash 0731 (max) (16.6%). It covers unpublished research-level physics problems. The board median is 2.9%, so Claude Sonnet 5 (max) is 14.0 points above it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.

On SciCode, Claude Sonnet 5 (max) reaches 53.6%. It ranks #26 of 106, ahead of 75% of the field, and 13.3 points behind Claude Opus 5.5 (max) at 66.9%. The results either side are Grok 4.5 (high) (54.1%) and GPT-5.6 Luna (max) (53.6%). It covers scientific research code. The board median is 44.3%, so Claude Sonnet 5 (max) is 9.3 points above it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.

Claude Sonnet 5 (max) manages 35.0% on Mystery Games, good for #17 of 59. GPT-6 Astra (max) leads at 84.0%, 49.0 points further up, and 71% of models sit below. The results either side are GPT-5.4 (xhigh) (37.0%) and GPT-5.6 Terra (max) (35.0%). It covers deductive reasoning puzzles. The board median is 23.0%, so Claude Sonnet 5 (max) is 12.0 points above it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.

Claude Sonnet 5 (max) lands below the middle on FrontierMath T4: 29.3%, #26 of 50, with 50% of the field scoring higher. GPT-6.1 Sol (max) leads at 100.0%; the closest result above is GPT-5.2 (xhigh) at 31.7%. It covers research-level maths, tier 4. The board median is 30.5%, so Claude Sonnet 5 (max) is 1.2 points below it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.

Claude Sonnet 5 (max) lands below the middle on SimpleQA Verified: 33.7%, #44 of 69, with 62% of the field scoring higher. gemini-4-argon-high leads at 76.0%; the closest result above is GLM-5.1 at 34.0%. It covers short factual questions. The board median is 41.0%, so Claude Sonnet 5 (max) is 7.3 points below it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.

DeepSWE is a weak spot. Claude Sonnet 5 (max) scores 53.8%, #16 of 19, behind 79% of the models measured. The leader, Gemini 3.8 Flash (high), reaches 73.8%. It covers long-horizon software engineering. The board median is 67.0%, so Claude Sonnet 5 (max) is 13.2 points below it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.

CursorBench is a weak spot. Claude Sonnet 5 (max) scores 34.1%, #14 of 15, behind 87% of the models measured. The leader, Claude Opus 5.5 (max), reaches 57.8%. It covers real-world coding-agent tasks. The board median is 41.6%, so Claude Sonnet 5 (max) is 7.5 points below it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.

The full table above also lists the other 4 results, including boards with fewer than ten models, where a rank says little.

Strengths and weaknesses

Software engineering: uneven, ahead of 33% of the field on average across 3 benchmarks. Its best board here is SciCode (#26 of 106) and its weakest is CursorBench (#14 of 15). Reasoning, maths and science: solid, ahead of 66% of the field on average across 6 benchmarks. Its best board here is ProofBench (#7 of 46) and its weakest is FrontierMath T4 (#26 of 50). General knowledge and chat: solid, ahead of 63% of the field on average across 4 benchmarks. Its best board here is LMCA (#14 of 108) and its weakest is SimpleQA Verified (#44 of 69).

Price and value

The median input price across the 427 priced models is $0.500 per million tokens, so Claude Sonnet 5 (max) costs more than the typical listing. Models that score within 1.5 points or better and cost less: GLM-5.2 (high) ($0.060, 51.6), DeepSeek V4 Flash 0731 (max) ($0.070, 53.1), Qwen3-Coder-30B-A3B-Instruct ($0.070, 52.7). Similar scores cost more on Grok 4 ($3.00) and claude-3.5-sonnet-20240620 ($3.00).

Release history

The Anthropic model released before it was Claude Fable 5 (max) on 2026-06-09, which scores 59.9. The next one was Claude Opus 5 (max) on 2026-07-24, at 60.0.

Reading these numbers

The 14 scores on this page come from 1 source: Epoch AI Benchmarking Hub (14). No official leaderboard supplied any of them, so treat them as third-party measurements. The leaderboard also lists 8 other reasoning-effort settings of this model, which redirect here. This page shows the highest.

Claude Sonnet 5 (max) is also listed at 8 other reasoning-effort settings; this page shows the highest-effort row, which is the one the leaderboard keeps by default.

Models ranked nearby

• #52 claude-3.5-sonnet-20240620 (52.7)
• #53 Qwen3-Coder-30B-A3B-Instruct (52.7)
• #54 GPT-5.1 (high) (52.7)
• #56 Claude 4.6 Opus (52.6)
• #57 Codex + GPT-6 Astra (52.6)
• #58 Codex + GPT-6.1 Sol (52.6)

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