BlogLLM Benchmark Scores as Published, Side by Side
Claude 3.5 Sonnet: Benchmark Scores as Published
Every result the catalog holds for Claude 3.5 Sonnet: 28 tests in 12 areas, from 3 pages. 10 printed by Anthropic itself, the rest by the company named beside each number. Each number is as printed; nothing is ranked or rescored.
Developer AnthropicReleased 2024-10-22Data version 2026103001, as of 2026-10-30
Agentic Coding
| Test | Value | Printed by | Setup as printed |
|---|---|---|---|
| SWE-bench Verified | 49% | Anthropic, Claude 3.7 Sonnetchecked 2026-10-15; printed on 1 more page | - |
Science and Math
| Test | Value | Printed by | Setup as printed |
|---|---|---|---|
| GPQA Diamond | 65% | Anthropic, Claude 3.7 Sonnetchecked 2026-10-15; printed on 1 more page | - |
| AIME 2024 | 16% | Anthropic, Claude 3.7 Sonnetchecked 2026-10-15; printed on 1 more page | - |
| MATH-500 | 78% | Anthropic, Claude 3.7 Sonnetchecked 2026-10-15; printed on 1 more page | - |
| CNMO 2024 | 13.1% | DeepSeek, DeepSeek-R1 model cardchecked 2026-10-15 | - |
| AIME 2024 cons@64 (majority vote over 64 samples) | 26.7% | DeepSeek, DeepSeek-R1 model cardchecked 2026-10-15 | - |
Computer Use
| Test | Value | Printed by | Setup as printed |
|---|---|---|---|
| OSWorld-Verified | 14.9% | Anthropic, Claude Sonnet 4.6checked 2026-10-15 | - |
Multimodal Understanding
| Test | Value | Printed by | Setup as printed |
|---|---|---|---|
| MMMU (validation) | 70.4% | Anthropic, Claude 3.7 Sonnetchecked 2026-10-15 | - |
General Knowledge and Reasoning
| Test | Value | Printed by | Setup as printed |
|---|---|---|---|
| MMLU-Pro | 78% | DeepSeek, DeepSeek-R1 model cardchecked 2026-10-15 | - |
| MMLU-Redux | 88.9% | DeepSeek, DeepSeek-R1 model cardchecked 2026-10-15 | - |
| C-Eval | 76.7% | DeepSeek, DeepSeek-R1 model cardchecked 2026-10-15 | - |
| MMLU | 88.3% | DeepSeek, DeepSeek-R1 model cardchecked 2026-10-15 | - |
| SimpleQA | 28.4% | DeepSeek, DeepSeek-R1 model cardchecked 2026-10-15 | Tools: none |
| DROP (F1) | 88.3% | DeepSeek, DeepSeek-R1 model cardchecked 2026-10-15 | - |
| Chinese SimpleQA | 55.4% | DeepSeek, DeepSeek-R1 model cardchecked 2026-10-15 | - |
Code Generation
| Test | Value | Printed by | Setup as printed |
|---|---|---|---|
| Codeforces rating | 717 | DeepSeek, DeepSeek-R1 model cardchecked 2026-10-15 | Setup as reported: metric Rating as printed in the label |
| 717 | DeepSeek, DeepSeek-R1 model cardchecked 2026-10-15 | Setup as reported: column of the distilled-model table; metric rating as printed in the header | |
| LiveCodeBench (version not stated) | 38.9% | DeepSeek, DeepSeek-R1 model cardchecked 2026-10-15 | Setup as reported: column of the distilled-model table; metric pass@1 as printed in the header |
| 33.8% | DeepSeek, DeepSeek-R1 model cardchecked 2026-10-15 | Setup as reported: metric Pass@1-COT as printed in the label | |
| Aider polyglot | 45.3% | DeepSeek, DeepSeek-R1 model cardchecked 2026-10-15 | - |
| Codeforces percentile | 20.3% | DeepSeek, DeepSeek-R1 model cardchecked 2026-10-15 | - |
Tool Use and Workflows
| Test | Value | Printed by | Setup as printed |
|---|---|---|---|
| τ-bench Airline | 48.8% | Anthropic, Claude 3.7 Sonnetchecked 2026-10-15 | - |
| τ-bench Retail | 71.5% | Anthropic, Claude 3.7 Sonnetchecked 2026-10-15 | - |
Web Search and Research
| Test | Value | Printed by | Setup as printed |
|---|---|---|---|
| FRAMES | 72.5% | DeepSeek, DeepSeek-R1 model cardchecked 2026-10-15 | - |
Instruction Following
| Test | Value | Printed by | Setup as printed |
|---|---|---|---|
| IFEval | 90.2% | Anthropic, Claude 3.7 Sonnetchecked 2026-10-15 | - |
| IFEval (prompt-level strict) | 86.5% | DeepSeek, DeepSeek-R1 model cardchecked 2026-10-15 | - |
Multilingual
| Test | Value | Printed by | Setup as printed |
|---|---|---|---|
| MMMLU | 82.1% | Anthropic, Claude 3.7 Sonnetchecked 2026-10-15 | - |
| CLUEWSC | 85.4% | DeepSeek, DeepSeek-R1 model cardchecked 2026-10-15 | - |
Writing and Conversation
| Test | Value | Printed by | Setup as printed |
|---|---|---|---|
| AlpacaEval 2.0 (length-controlled win rate) | 52% | DeepSeek, DeepSeek-R1 model cardchecked 2026-10-15 | - |
Preference and Composite Rankings
| Test | Value | Printed by | Setup as printed |
|---|---|---|---|
| Arena-Hard win rate (GPT-4-1106 judge) | 85.2% | DeepSeek, DeepSeek-R1 model cardchecked 2026-10-15 | - |
More from Anthropic
- Claude Haiku 5.5
- Claude Sonnet 5.5
- Claude Opus 5.5
- Claude Fable 5.1
- Claude Mythos 5.1
- Claude Opus 5
- Claude Sonnet 5
- Claude Fable 5
- Claude Mythos 5
- Claude Opus 4.8
- Claude Opus 4.7
- Claude Mythos Preview
- Claude Sonnet 4.6
- Claude Opus 4.6
- Claude Opus 4.5
- Claude Haiku 4.5
- Claude Sonnet 4.5
- Claude Opus 4.1
- Claude Opus 4
- Claude Sonnet 4
- Claude 3.7 Sonnet
Every number here is read from the same catalog file as the benchmark table, data version 2026103001. Where one page printed the model under several setups, each value is listed. The catalog is public in armanobosyan/sugra-llm-benchmarks; a wrong number can be reported from the table, with Report an error under the value, or as an issue in that repository.