Claude Sonnet 3.5 v2
| Released | 2024-10-22 |
|---|---|
| Price per 1M | $3.00 in · $15.00 out |
| Overall score | 41.8 · rank 186 |
| Context window | 200,000 tokens |
| Provider | Anthropic |
| Weights | Proprietary |
Scores
| Dimension | Score | Rank | Benchmarks |
|---|---|---|---|
| Overall: Balanced | 41.8 | 186 | 20 |
| Overall: Coding | 42.3 | 176 | 8 |
| Overall: Reasoning | 39.9 | 194 | 9 |
| Overall: Agentic work | 41.0 | 184 | 8 |
| Overall: Frontend | 43.3 | 153 | 7 |
| Coding | 43.2 | 167 | 3 |
| Agents | 39.9 | 155 | 2 |
| Reasoning | 42.0 | 174 | 3 |
| Math | 39.1 | 165 | 4 |
| Knowledge | 36.7 | 151 | 2 |
| General | 48.3 | 125 | 2 |
| Multimodal | 43.8 | 94 | 4 |
| Value vs price | −11.3 | 180 |
Evidence
Coding 43.2 ±3.9
| Benchmark | Result | Source | Date |
|---|---|---|---|
| GSO | 4.6% | Epoch AI | 2026-11-11 |
| Aider Polyglot | 51.6% | Epoch AI | 2026-11-11 |
| LiveCodeBench (Vals run) | 49.6% | Vals AI | 2026-09-01 |
Agents 39.9 ±3.6
| Benchmark | Result | Source | Date |
|---|---|---|---|
| METR 50% Time Horizon | $21 | METR | 2026-11-11 |
| METR task suite | 45.2% | Epoch AI | 2026-11-11 |
| BALROG | 32.6% | Epoch AI | 2026-11-11 |
Reasoning 42.0 ±4.2
| Benchmark | Result | Source | Date |
|---|---|---|---|
| EnigmaEval | 0.9% | Epoch AI | 2026-11-11 |
| SimpleBench | 41.4% | Epoch AI | 2026-11-11 |
| GPQA Diamond | 55.3% | Epoch AI | 2025-01-27 |
| GPQA Diamond (Vals run) | 59.3% | Vals AI | 2026-09-01 |
Math 39.1 ±3.0
| Benchmark | Result | Source | Date |
|---|---|---|---|
| MATH Level 5 | 57.0% | Epoch AI | 2025-01-27 |
| FrontierMath Tiers 1-3 (2025 set) | 2.1% | Epoch AI | 2025-03-06 |
| FrontierMath Tier 4 (2025 set) | 0.0% | Epoch AI | 2025-07-01 |
| OTIS Mock AIME 2024-2025 | 8.5% | Epoch AI | 2025-02-25 |
Knowledge 36.7 ±1.3
| Benchmark | Result | Source | Date |
|---|---|---|---|
| Humanity's Last Exam | 4.1% | Epoch AI | 2026-11-11 |
| MMLU-Pro (Vals run) | 78.4% | Vals AI | 2026-09-01 |
General 48.3 ±2.6
| Benchmark | Result | Source | Date |
|---|---|---|---|
| EQ-Bench Creative Writing v3 | 1,451 | EQ-Bench | 2026-11-11 |
| LMArena Text | 1,374 | LMArena | 2026-11-09 |
Multimodal 43.8 ±2.8
| Benchmark | Result | Source | Date |
|---|---|---|---|
| LMArena Vision | 1,161 | LMArena | 2026-11-09 |
| MMMU-Pro (Vals run) | 68.8% | Vals AI | 2026-09-01 |
| VPCT | 33.0% | Epoch AI | 2026-11-11 |
| GeoBench | 62.0% | Epoch AI | 2026-11-11 |
Names matched to this model
claude-sonnet-3-5-20241022, claude-sonnet-3-5-v2