AI BENCHY Compare
DeepSeek: DeepSeek V3.2 vs StepFun: Step 3.5 Flash
Benchmarks gegenereerd uit AI BENCHY-testsuites op: 2026-03-03
| Metriek | DeepSeek: DeepSeek V3.2 medium Releasedatum: 2025-12-01 | StepFun: Step 3.5 Flash medium Releasedatum: 2026-02-01 Gratis beschikbaar |
|---|---|---|
| Rang | #12 | #11 |
| Gem. score | 6.98 | 7.00 |
| Consistentie | 8.75 | 8.32 |
| Kosten per resultaat | 0.193 | 0.000 |
| Totale kosten | $0.018 | $0.000 |
| Correcte tests | 9/14 | 9/14 |
| Slaagpercentage per poging | 71.4% | 73.8% |
| Instabiele tests | 2 | 3 |
| Uitvoer-tokens | 6,753 | 60,502 |
| Redeneer-tokens | 30,427 | 117,044 |
Categorie-uitsplitsing
| Anti-AI-trucs | Score | Consistentie | Slaagpercentage per poging | Instabiele tests | Correcte tests | Uitvoer-tokens | Redeneer-tokens |
|---|---|---|---|---|---|---|---|
| DeepSeek: DeepSeek V3.2 | 7.00 | 9.86 | 66.7% | 0 | 1,171 | 4,893 | |
| StepFun: Step 3.5 Flash | 10.00 | 10.00 | 100.0% | 0 | 13,924 | 17,208 |
| Gegevensparsering en extractie | Score | Consistentie | Slaagpercentage per poging | Instabiele tests | Correcte tests | Uitvoer-tokens | Redeneer-tokens |
|---|---|---|---|---|---|---|---|
| DeepSeek: DeepSeek V3.2 | 9.88 | 10.00 | 100.0% | 0 | 207 | 7,693 | |
| StepFun: Step 3.5 Flash | 10.00 | 10.00 | 100.0% | 0 | 535 | 11,548 |
| Domeinspecifiek | Score | Consistentie | Slaagpercentage per poging | Instabiele tests | Correcte tests | Uitvoer-tokens | Redeneer-tokens |
|---|---|---|---|---|---|---|---|
| DeepSeek: DeepSeek V3.2 | 4.00 | 7.21 | 44.4% | 1 | 3,081 | 7,856 | |
| StepFun: Step 3.5 Flash | 4.00 | 7.21 | 44.4% | 1 | 40,942 | 74,237 |
| Instructies opvolgen | Score | Consistentie | Slaagpercentage per poging | Instabiele tests | Correcte tests | Uitvoer-tokens | Redeneer-tokens |
|---|---|---|---|---|---|---|---|
| DeepSeek: DeepSeek V3.2 | 7.00 | 9.84 | 50.0% | 0 | 1,397 | 2,845 | |
| StepFun: Step 3.5 Flash | 10.00 | 10.00 | 100.0% | 0 | 2,121 | 3,274 |
| Puzzle Solving | Score | Consistentie | Slaagpercentage per poging | Instabiele tests | Correcte tests | Uitvoer-tokens | Redeneer-tokens |
|---|---|---|---|---|---|---|---|
| DeepSeek: DeepSeek V3.2 | 7.00 | 7.21 | 88.9% | 1 | 390 | 6,281 | |
| StepFun: Step 3.5 Flash | 2.00 | 4.96 | 33.3% | 2 | 2,705 | 6,975 |
| Toolaanroepen | Score | Consistentie | Slaagpercentage per poging | Instabiele tests | Correcte tests | Uitvoer-tokens | Redeneer-tokens |
|---|---|---|---|---|---|---|---|
| DeepSeek: DeepSeek V3.2 | 10.00 | 10.00 | 100.0% | 0 | 507 | 859 | |
| StepFun: Step 3.5 Flash | 10.00 | 10.00 | 100.0% | 0 | 275 | 3,802 |
Snelle vergelijking
Vergelijkingspaar wisselen
Gemini 3 Flash PreviewnonevsStep 3.5 FlashmediumGratis beschikbaarDeepSeek V3.2mediumvsGemini 3 Flash PreviewnoneClaude Sonnet 4.6nonevsDeepSeek V3.2mediumClaude Sonnet 4.6nonevsStep 3.5 FlashmediumGratis beschikbaarDeepSeek V3.2mediumvsQwen3.5 Plus 2026-02-15noneQwen3.5 Plus 2026-02-15nonevsStep 3.5 FlashmediumGratis beschikbaarDeepSeek V3.2mediumvsGLM 5noneStep 3.5 FlashmediumGratis beschikbaarvsGLM 5noneGemini 3 Flash PreviewlowvsStep 3.5 FlashmediumGratis beschikbaarDeepSeek V3.2mediumvsGemini 3 Flash PreviewlowDeepSeek V3.2mediumvsQwen3.5-122B-A10BnoneQwen3.5-122B-A10BnonevsStep 3.5 FlashmediumGratis beschikbaar