Benchmark Almanac

InternVL3-14B

InternVL3-14B places in the top three on ViewSpatial-Bench.

Shanghai AI Lab Released April 11th, 2025 Family InternVL3 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
Shanghai AI Lab Apr 2025

InternVL3-14B

Open weights Mid-size

No Epoch Capabilities Index score yet
  • Video SimpleQA 25.2% #13 /27
  • PhyX 7.90% #7 /10
  • ViewSpatial-Bench 40.3% #2 /7

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 83.3% #26 of 178 Official leaderboard
Video SimpleQA 25.2% #13 of 27 Official leaderboard

Math and science

BenchmarkScoreRankSource
PhyX 7.90% #7 of 10 Official leaderboard

Coding

BenchmarkScoreRankSource
SArena 0.82 #10 of 17 Official leaderboard

Vision, audio and robotics

BenchmarkScoreRankSource
MM-Vet 80.5% #4 of 164 Official leaderboard
MMBench 85.7% #15 of 155 Official leaderboard
MMStar 68.9% #17 of 173 Official leaderboard
CCBench (MMBench Chinese Culture) 77.1% #15 of 152 Official leaderboard
SEED-Bench-2-Plus 70.1% #16 of 136 Official leaderboard
MMBench-EN v1.1 83.9% #25 of 178 Official leaderboard
CharXiv (DQ) 82.2% #12 of 67 Official leaderboard
A-Bench 76.3% #23 of 116 Official leaderboard
CharXiv 43.1% #16 of 77 Official leaderboard
SpaCE-10 47.3% #10 of 36 Official leaderboard
ViewSpatial-Bench 40.3% #2 of 7 Official leaderboard
MMMU 67.1% #42 of 136 Official leaderboard
Ego3D-Bench 7.70 #4 of 11 Official leaderboard
ERQA 38.8% #7 of 17 Official leaderboard
MATH-Vision (MathVision) 37% #65 of 105 Official leaderboard

Other

BenchmarkScoreRankSource
OpenVLM Leaderboard 75.2 #12 of 173 Official leaderboard