Benchmark Almanac

CogVLM-17B-Chat

CogVLM-17B-Chat places in the top three on VisualWebBench.

Z.ai Released November 16th, 2023 Family CogVLM The lab's model card Compare with others
20
Benchmarks with a score
0
Ranked first, out of five or more models
1
In the top three
Z.ai Nov 2023

CogVLM-17B-Chat

Open weights Small

No Epoch Capabilities Index score yet
  • MM-Vet 54.5% #72 /164
  • SEED-Bench-2-Plus 54.6% #95 /136
  • CoreCognition 49.7% #96 /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 54% #158 of 178 Official leaderboard
AMBER 83.4 #2 of 2 Official leaderboard

Agents and tools

BenchmarkScoreRankSource
VisualWebBench 25.1% #3 of 7 Official leaderboard
VisualWebArena 0.33% #7 of 7 Official leaderboard

Professional work

BenchmarkScoreRankSource
GMAI-MMBench 36.1% #10 of 13 Official leaderboard

Vision, audio and robotics

BenchmarkScoreRankSource
VCR (Visual Caption Restoration) 35.3% #11 of 42 Official leaderboard
MM-Vet 54.5% #72 of 164 Official leaderboard
SEED-Bench-2-Plus 54.6% #95 of 136 Official leaderboard
A-Bench 66.9% #91 of 116 Official leaderboard
CoreCognition 49.7% #96 of 119 Official leaderboard
BLINK 39.4% #9 of 11 Official leaderboard
MMStar 39.9% #142 of 173 Official leaderboard
MMBench 65.8% #130 of 155 Official leaderboard
MMBench-EN v1.1 63.6% #150 of 178 Official leaderboard
MMT-Bench 51.6% #13 of 15 Official leaderboard
CCBench (MMBench Chinese Culture) 19% #138 of 152 Official leaderboard
MME 1439.1 #16 of 17 Official leaderboard
MMMU 32.1% #134 of 136 Official leaderboard
MuirBench 20.9% #8 of 8 Official leaderboard

Other

BenchmarkScoreRankSource
OpenVLM Leaderboard 47.9 #134 of 173 Official leaderboard