Side-by-side benchmark comparison across knowledge, coding, math, and reasoning.
Llama 3.1 405B wins overall with a score of 58 vs 48 (10 point difference).Llama 3.1 405B wins 4 out of 4 categories.
Llama 3.1 405B
68.5
Llama 3 70B
56.5
Llama 3.1 405B
62
Llama 3 70B
50
Llama 3.1 405B
69
Llama 3 70B
57
Llama 3.1 405B
67
Llama 3 70B
55
Llama 3.1 405B scores higher overall with 58 vs 48, a difference of 10 points across all benchmarks.
Llama 3.1 405B leads in knowledge tasks with an average score of 68.5 vs 56.5.
Llama 3.1 405B leads in coding with an average score of 62 vs 50.
Llama 3.1 405B leads in math with an average score of 69 vs 57.
Llama 3.1 405B leads in reasoning with an average score of 67 vs 55.