Benchmark Almanac

Llama 3.2 11B

Llama 3.2 11B.

Meta Released 24 Sep 2024
26
Benchmarks with a score
0
Ranked first, out of five or more models
0
In the top three
Meta Sep 2024

Llama 3.2 11B

Open weights

No Epoch Capabilities Index score yet
  • CritPt 0% #116 /160
  • VisualPuzzles 29.4% #24 /29
  • HumanEval-V 1.60% #19 /24

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
Vectara Hallucination Leaderboard (HHEM) 5.50% #48 of 148 Official leaderboard
EQ-Bench v2 67.9 #58 of 85 Official leaderboard
MMLU 56.5% #91 of 127 Epoch AI
MMBench-CN v1.1 64.2% #129 of 178 Official leaderboard

Math and science

BenchmarkScoreRankSource
CritPt 0% #116 of 160 Epoch AI
U-MATH 20.4% #30 of 30 Official leaderboard

Coding

BenchmarkScoreRankSource
HumanEval-V 1.60% #19 of 24 Official leaderboard

Agents and tools

BenchmarkScoreRankSource
BALROG 16.8% #27 of 35 Official leaderboard
SnakeBench 18 #146 of 183 Official leaderboard

Professional work

BenchmarkScoreRankSource
MortgageTax 23.3% #85 of 86 Official leaderboard

Vision, audio and robotics

BenchmarkScoreRankSource
CharXiv 31.2% #43 of 77 Official leaderboard
MMStar 49.8% #105 of 173 Official leaderboard
MMMU 50.7% #83 of 136 Official leaderboard
A-Bench 69.2% #77 of 116 Official leaderboard
MMVU 38.9% #20 of 28 Official leaderboard
CCBench (MMBench Chinese Culture) 30.2% #111 of 152 Official leaderboard
MMBench-EN v1.1 67.5% #134 of 178 Official leaderboard
MMMU-Pro 28.4% #44 of 57 Official leaderboard
MMBench 68% #121 of 155 Official leaderboard
VisualPuzzles 29.4% #24 of 29 Official leaderboard
MEGA-Bench 18% #19 of 22 Official leaderboard
Video-MMMU 30% #21 of 23 Official leaderboard
Vision Arena 993 #131 of 136 Official leaderboard
VISTA 20.5% #53 of 54 Official leaderboard

Other

BenchmarkScoreRankSource
OpenVLM Leaderboard 57.7 #95 of 173 Official leaderboard
HELM Lite 32.5 #57 of 71 Official leaderboard