Benchmark Almanac

Janus-1.3B

Janus-1.3B.

DeepSeek Released 18 Oct 2024 Family Janus The lab's model card
8
Benchmarks with a score
0
Ranked first, out of five or more models
0
In the top three
DeepSeek Oct 2024

Janus-1.3B

Open weights Small

No Epoch Capabilities Index score yet
  • OCRBench v2 14.3% #65 /65
  • A-Bench 59.1% #111 /116
  • MMBench-EN v1.1 50.2% #168 /178

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 50.4% #162 of 178 Official leaderboard

Vision, audio and robotics

BenchmarkScoreRankSource
CCBench (MMBench Chinese Culture) 22.2% #134 of 152 Official leaderboard
MMStar 37.6% #158 of 173 Official leaderboard
MMBench-EN v1.1 50.2% #168 of 178 Official leaderboard
MMBench 52% #148 of 155 Official leaderboard
A-Bench 59.1% #111 of 116 Official leaderboard
OCRBench v2 14.3% #65 of 65 Official leaderboard

Other

BenchmarkScoreRankSource
OpenVLM Leaderboard 40.2 #159 of 173 Official leaderboard