MVBench
Resultados de evaluaciones comparativas de IA y rendimiento de modelos
Tipo: percentageMultimodal
Resultados registrados
40
Puntuación media
70.68%
Rango de puntuaciones
68.70% - 76.60%
Modelo líder
76.60% - Qwen 3.5 122B A10B
Puntuaciones a lo largo del tiempo
Puntuaciones individuales de la evaluación a lo largo del tiempo.
Resultados de modelos
| Organización | Modelos | Fecha del resultado | Mejor puntuación | Información | Autodeclarado | Fuente |
|---|---|---|---|---|---|---|
| Qwen 3.5 122B A10B | 24 feb 2026 | 76,60 % | - | Sí | Fuente | |
| Qwen 3.5 35B A3B | 24 feb 2026 | 74,80 % | - | Sí | Fuente | |
| Qwen 3.5 27B | 24 feb 2026 | 74,60 % | - | Sí | Fuente | |
| Qwen 3.5 Flash | 23 feb 2026 | 74,60 % | inferred family alias from qwen3.5-27b (score=0.4147; benches=81) | Sí | Fuente | |
| Qwen 2 VL 2B | - | 73,60 % | inferred family alias from qwen2-vl-72b (score=0.4083; benches=15) | Sí | Fuente | |
| Qwen 2 VL 72B | - | 73,60 % | - | Sí | Fuente | |
| Qwen 3 VL 32B Thinking | - | 73,20 % | - | Sí | - | |
| Qwen 3 VL 32B Instruct | - | 72,80 % | - | Sí | - | |
| Qwen 3 VL 30B A3B Instruct | - | 72,30 % | - | Sí | - | |
| Qwen 3 VL 30B A3B Thinking | - | 72 % | - | Sí | - | |
| Qwen 2.5 Math 72B | - | 70,40 % | inferred high-confidence family alias from qwen2.5-vl-72b (score=0.4700; benches=30) | Sí | Fuente | |
| Qwen 2.5 Math PRM 72B | - | 70,40 % | inferred family alias from qwen2.5-vl-72b (score=0.4050; benches=30) | Sí | Fuente | |
| Qwen 2.5 Math RM 72B | - | 70,40 % | inferred family alias from qwen2.5-vl-72b (score=0.4092; benches=30) | Sí | Fuente | |
| Qwen 2.5 VL 72B Instruct | 26 ene 2025 | 70,40 % | - | Sí | Fuente | |
| Qwen 2.5 Coder 3B | - | 70,30 % | inferred family alias from qwen2.5-omni-7b (score=0.3000; benches=45) | Sí | Fuente | |
| Qwen 2.5 Coder 7B | - | 70,30 % | inferred high-confidence family alias from qwen2.5-omni-7b (score=0.4700; benches=45) | Sí | Fuente | |
| Qwen 2.5 Math 7B | - | 70,30 % | inferred high-confidence family alias from qwen2.5-omni-7b (score=0.4767; benches=45) | Sí | Fuente | |
| Qwen 2.5 Math 7B PRM800K | - | 70,30 % | inferred family alias from qwen2.5-omni-7b (score=0.3696; benches=45) | Sí | Fuente | |
| Qwen 2.5 Math PRM 7B | - | 70,30 % | inferred family alias from qwen2.5-omni-7b (score=0.4092; benches=45) | Sí | Fuente | |
| Qwen 2.5 Omni 3B | - | 70,30 % | inferred high-confidence family alias from qwen2.5-omni-7b (score=0.4933; benches=45) | Sí | Fuente | |
| Qwen 2.5 Omni 7B | - | 70,30 % | - | Sí | Fuente | |
| Qwen 2.5 0.5B | - | 69,60 % | inferred family alias from qwen2.5-vl-7b (score=0.3131; benches=32) | Sí | Fuente | |
| Qwen 2.5 1.5B | - | 69,60 % | inferred family alias from qwen2.5-vl-7b (score=0.3131; benches=32) | Sí | Fuente | |
| Qwen 2.5 3B | - | 69,60 % | inferred family alias from qwen2.5-vl-7b (score=0.3531; benches=32) | Sí | Fuente | |
| Qwen 2.5 VL 7B Instruct | - | 69,60 % | - | Sí | Fuente |