Benchmark Almanac

llava-v1.6-mistral-7b

llava-v1.6-mistral-7b places in the top three on 3 benchmarks.

Released January 31st, 2024 Compare with others
22
Benchmarks with a score
0
Ranked first, out of five or more models
3
In the top three
Unknown lab Jan 2024

llava-v1.6-mistral-7b

No Epoch Capabilities Index score yet
  • MM-Vet 42.2% #113 /164
  • SEED-Bench-2-Plus 49.3% #108 /136
  • CoreCognition 55.8% #60 /119

Where it ranks

Its best score on each benchmark, and where that ranks among every model with a score there. We rank each model once, at its best setting.

Knowledge and language

BenchmarkScoreRankSource
MMBench-CN v1.1 59.8% #141 of 178 Official leaderboard

Coding

BenchmarkScoreRankSource
ChartMimic 21.0 #11 of 14 Official leaderboard

Vision, audio and robotics

BenchmarkScoreRankSource
MMDU 42.8 #2 of 5 Official leaderboard
ScienceQA 72.8% #7 of 17 Claim in Falcon2-11B Technical Report, via Epoch AI
ChartQA Pro 22% #3 of 7 Official leaderboard
NaturalBench 16.3% #3 of 6 Official leaderboard
CoreCognition 55.8% #60 of 119 Official leaderboard
VCR (Visual Caption Restoration) 18.3% #28 of 42 Official leaderboard
Q-Bench-Video 47.5% #2 of 3 Official leaderboard
MM-Vet 42.2% #113 of 164 Official leaderboard
A-Bench 68.3% #84 of 116 Official leaderboard
MMBench 69.6% #114 of 155 Official leaderboard
CCBench (MMBench Chinese Culture) 30% #112 of 152 Official leaderboard
LongVideoBench 47.1% #17 of 23 Official leaderboard
MMBench-EN v1.1 66.5% #139 of 178 Official leaderboard
SEED-Bench-2-Plus 49.3% #108 of 136 Official leaderboard
MMStar 38.4% #150 of 173 Official leaderboard
CharXiv (DQ) 35.4% #62 of 67 Official leaderboard
MMMU 35.3% #130 of 136 Official leaderboard
MMMU-Pro 17% #56 of 57 Official leaderboard
CharXiv 13.9% #76 of 77 Official leaderboard

Other

BenchmarkScoreRankSource
OpenVLM Leaderboard 45.9 #145 of 173 Official leaderboard