Benchmark RadarAI BENCHMARK PROFILE
MMVetGPT4Turbo
MM-Vet evaluation using GPT-4 Turbo for scoring. This variant of MM-Vet examines large multimodal models on complicated multimodal tasks requiring integrated capabilities across six core vision-language abilities: recognition, knowledge, spatial awareness, language generation, OCR, and math.
- Released
- Unknown
- Readiness
- Paper only
- Primary field
- General AI
Primary resources
Benchmark Radar records only publicly supported details and links back to primary sources for verification.