Back to models

Llama 3-8B

ECI#209
116
Meta AI
Apr. 18, 2024
Open weights

Llama 3-8B is an open-weights model from Meta AI, released April 18, 2024. It has an ECI score of 116 and ranks 209th of 253 tracked models. At the time of its release, it ranked 19th of 56 tracked models.

Llama 3-8B vs select alternatives

DomainBenchmark
Llama 3-8B
Meta AI
Apr. 18, 2024
Open weights
Meta AI
Jul. 18, 2023
Open weights
Microsoft
Apr. 23, 2024
Open weights
Anthropic
Mar. 7, 2024
Closed weights
ECI scoreBest score 167
116114117118
Ranking#209#212#205#202
Conceptual reasoning
DTBenchBest score 99%
44%
42%
n/a
50%
Games
Chess PuzzlesBest score 72%
0%
n/a
0%
n/a
Language understanding
Adversarial NLIBest score 58%
57%
n/a
53%
n/a
Long context
ForecastBenchBest score 62.5
58.651.4—53.2
Mathematics
MATH Level 5Best score 98%
6%
3%
n/a
15%
OTIS Mock AIME 2024-2025Best score 100%
2%
0%
n/a
2%
Science
ARC AI2Best score 95%
83%
n/a
85%
n/a
GPQA DiamondBest score 96%
26%
26%
n/a
36%
World knowledge
MMLUBest score 88%
69%
60%
69%
74%

Model metadata