AI BENCHY Compare
Anthropic: Claude Sonnet 4.6 vs DeepSeek: DeepSeek V3.2
Benchmarks gegenereerd uit AI BENCHY-testsuites op: 2026-03-03
| Metriek | Anthropic: Claude Sonnet 4.6 none Releasedatum: 2026-02-17 | DeepSeek: DeepSeek V3.2 medium Releasedatum: 2025-12-01 |
|---|---|---|
| Rang | #15 | #12 |
| Gem. score | 6.36 | 6.98 |
| Consistentie | 9.50 | 8.75 |
| Kosten per resultaat | 1.325 | 0.193 |
| Totale kosten | $0.106 | $0.018 |
| Correcte tests | 8/14 | 9/14 |
| Slaagpercentage per poging | 59.5% | 71.4% |
| Instabiele tests | 1 | 2 |
| Uitvoer-tokens | 2,637 | 6,753 |
| Redeneer-tokens | 0 | 30,427 |
Categorie-uitsplitsing
| Anti-AI-trucs | Score | Consistentie | Slaagpercentage per poging | Instabiele tests | Correcte tests | Uitvoer-tokens | Redeneer-tokens |
|---|---|---|---|---|---|---|---|
| Anthropic: Claude Sonnet 4.6 | 4.00 | 10.00 | 33.3% | 0 | 1,199 | 0 | |
| DeepSeek: DeepSeek V3.2 | 7.00 | 9.86 | 66.7% | 0 | 1,171 | 4,893 |
| Gegevensparsering en extractie | Score | Consistentie | Slaagpercentage per poging | Instabiele tests | Correcte tests | Uitvoer-tokens | Redeneer-tokens |
|---|---|---|---|---|---|---|---|
| Anthropic: Claude Sonnet 4.6 | 10.00 | 10.00 | 100.0% | 0 | 252 | 0 | |
| DeepSeek: DeepSeek V3.2 | 9.88 | 10.00 | 100.0% | 0 | 207 | 7,693 |
| Domeinspecifiek | Score | Consistentie | Slaagpercentage per poging | Instabiele tests | Correcte tests | Uitvoer-tokens | Redeneer-tokens |
|---|---|---|---|---|---|---|---|
| Anthropic: Claude Sonnet 4.6 | 7.00 | 10.00 | 66.7% | 0 | 54 | 0 | |
| DeepSeek: DeepSeek V3.2 | 4.00 | 7.21 | 44.4% | 1 | 3,081 | 7,856 |
| Instructies opvolgen | Score | Consistentie | Slaagpercentage per poging | Instabiele tests | Correcte tests | Uitvoer-tokens | Redeneer-tokens |
|---|---|---|---|---|---|---|---|
| Anthropic: Claude Sonnet 4.6 | 5.50 | 10.00 | 50.0% | 0 | 90 | 0 | |
| DeepSeek: DeepSeek V3.2 | 7.00 | 9.84 | 50.0% | 0 | 1,397 | 2,845 |
| Puzzle Solving | Score | Consistentie | Slaagpercentage per poging | Instabiele tests | Correcte tests | Uitvoer-tokens | Redeneer-tokens |
|---|---|---|---|---|---|---|---|
| Anthropic: Claude Sonnet 4.6 | 5.00 | 7.68 | 44.4% | 1 | 595 | 0 | |
| DeepSeek: DeepSeek V3.2 | 7.00 | 7.21 | 88.9% | 1 | 390 | 6,281 |
| Toolaanroepen | Score | Consistentie | Slaagpercentage per poging | Instabiele tests | Correcte tests | Uitvoer-tokens | Redeneer-tokens |
|---|---|---|---|---|---|---|---|
| Anthropic: Claude Sonnet 4.6 | 10.00 | 10.00 | 100.0% | 0 | 447 | 0 | |
| DeepSeek: DeepSeek V3.2 | 10.00 | 10.00 | 100.0% | 0 | 507 | 859 |
Snelle vergelijking
Vergelijkingspaar wisselen
Claude Sonnet 4.6nonevsGrok 4.1 FastmediumClaude Sonnet 4.6nonevsKimi K2.5mediumDeepSeek V3.2mediumvsGemini 3 Flash PreviewnoneClaude Sonnet 4.6nonevsSeed-2.0-MinimediumClaude Sonnet 4.6nonevsMiMo-V2-FlashmediumClaude Sonnet 4.6nonevsQwen3.5-122B-A10BmediumClaude Sonnet 4.6nonevsQwen3.5-35B-A3BmediumClaude Sonnet 4.6nonevsGPT-5 NanomediumClaude Sonnet 4.6nonevsQwen3.5-FlashmediumClaude Sonnet 4.6nonevsGPT-5.2mediumClaude Sonnet 4.6nonevsStep 3.5 FlashmediumGratis beschikbaarDeepSeek V3.2mediumvsQwen3.5 Plus 2026-02-15none