AI BENCHY Compare
DeepSeek: DeepSeek V3.2 vs MiniMax: MiniMax M2.5
Benchmarks gegenereerd uit AI BENCHY-testsuites op: 2026-03-03
| Metriek | DeepSeek: DeepSeek V3.2 none Releasedatum: 2025-12-01 | MiniMax: MiniMax M2.5 medium Releasedatum: 2026-02-12 |
|---|---|---|
| Rang | #30 | #27 |
| Gem. score | 4.70 | 5.64 |
| Consistentie | 8.19 | 6.12 |
| Kosten per resultaat | 0.132 | 4.028 |
| Totale kosten | $0.007 | $0.242 |
| Correcte tests | 5/14 | 6/14 |
| Slaagpercentage per poging | 47.6% | 64.3% |
| Instabiele tests | 3 | 7 |
| Uitvoer-tokens | 4,869 | 121,297 |
| Redeneer-tokens | 0 | 203,513 |
Categorie-uitsplitsing
| Anti-AI-trucs | Score | Consistentie | Slaagpercentage per poging | Instabiele tests | Correcte tests | Uitvoer-tokens | Redeneer-tokens |
|---|---|---|---|---|---|---|---|
| DeepSeek: DeepSeek V3.2 | 1.00 | 9.68 | 0.0% | 0 | 1,411 | 0 | |
| MiniMax: MiniMax M2.5 | 9.33 | 7.89 | 88.9% | 1 | 286 | 45,112 |
| Gegevensparsering en extractie | Score | Consistentie | Slaagpercentage per poging | Instabiele tests | Correcte tests | Uitvoer-tokens | Redeneer-tokens |
|---|---|---|---|---|---|---|---|
| DeepSeek: DeepSeek V3.2 | 5.38 | 5.81 | 66.7% | 1 | 1,710 | 0 | |
| MiniMax: MiniMax M2.5 | 5.50 | 5.81 | 83.3% | 1 | 369 | 4,952 |
| Domeinspecifiek | Score | Consistentie | Slaagpercentage per poging | Instabiele tests | Correcte tests | Uitvoer-tokens | Redeneer-tokens |
|---|---|---|---|---|---|---|---|
| DeepSeek: DeepSeek V3.2 | 1.00 | 7.21 | 22.2% | 1 | 24 | 0 | |
| MiniMax: MiniMax M2.5 | 1.00 | 4.41 | 22.2% | 2 | 111,023 | 139,533 |
| Instructies opvolgen | Score | Consistentie | Slaagpercentage per poging | Instabiele tests | Correcte tests | Uitvoer-tokens | Redeneer-tokens |
|---|---|---|---|---|---|---|---|
| DeepSeek: DeepSeek V3.2 | 8.00 | 9.99 | 50.0% | 0 | 66 | 0 | |
| MiniMax: MiniMax M2.5 | 7.00 | 6.41 | 66.7% | 1 | 1,121 | 2,521 |
| Puzzle Solving | Score | Consistentie | Slaagpercentage per poging | Instabiele tests | Correcte tests | Uitvoer-tokens | Redeneer-tokens |
|---|---|---|---|---|---|---|---|
| DeepSeek: DeepSeek V3.2 | 7.67 | 7.49 | 88.9% | 1 | 1,136 | 0 | |
| MiniMax: MiniMax M2.5 | 4.33 | 4.79 | 55.6% | 2 | 8,229 | 10,458 |
| 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 | 522 | 0 | |
| MiniMax: MiniMax M2.5 | 10.00 | 10.00 | 100.0% | 0 | 269 | 937 |
Snelle vergelijking
Vergelijkingspaar wisselen
MiniMax M2.5mediumvsGLM 5noneMiniMax M2.5mediumvsQwen3.5-122B-A10BnoneDeepSeek V3.2nonevsGLM 4.7 FlashmediumMiniMax M2.5mediumvsQwen3.5 Plus 2026-02-15noneClaude Sonnet 4.6nonevsMiniMax M2.5mediumMiniMax M2.5mediumvsGPT-4o-mininoneDeepSeek V3.2nonevsgpt-oss-120bmediumGratis beschikbaarMiniMax M2.5mediumvsQwen3.5-27BnoneMiniMax M2.5mediumvsQwen3.5-35B-A3BnoneDeepSeek V3.2nonevsQwen3.5-FlashmediumDeepSeek V3.2nonevsGPT-5 NanomediumDeepSeek V3.2nonevsQwen3 Coder Nextmedium