Live
Head to head

GPT-4 (Mar 2023) vs Llama 3.1-405B

OpenAI
GPT-4 (Mar 2023)
March 2023
vs
Meta AI
Llama 3.1-405B
July 2024
2.1×10²⁵Training compute (FLOP)3.8×10²⁵
125.9Capability index129
$37MTraining cost$53M
Benchmarks won (1 scored on both)
0 GPT-4 (Mar 2023)Llama 3.1-405B 1

GPT-4 (Mar 2023) (OpenAI) and Llama 3.1-405B (Meta AI) are both frontier AI models. GPT-4 (Mar 2023) was published in March 2023 and Llama 3.1-405B in July 2024.

Llama 3.1-405B was trained on 3.8×10²⁵ FLOP, about 1.8x the compute of GPT-4 (Mar 2023) at 2.1×10²⁵ FLOP. Training compute is the closest available proxy for how much was invested in a model, though it says nothing on its own about how well that compute was spent.

Llama 3.1-405B wins the single benchmark both were scored on. Benchmark counts are a crude scoreboard — the evaluations differ wildly in what they measure and in how saturated they are — so the per-benchmark table below matters more than the tally.

Specifications
OpenAI
Organization
Meta AI
Mar 15, 2023
Published
Jul 23, 2024
2.1×10²⁵ FLOP
Training compute1.8x
3.8×10²⁵ FLOP
1.8T
Parameters4.4x
405B
5.4T
Dataset size2.9x
15.6T
NVIDIA A100 SXM4 40 GB
Training hardware
NVIDIA H100 SXM5 80GB
25,000
Chips used1.5x
16,384
2,280 h
Training time
2,142 h
$37M
Training cost (2023 USD)1.4x
$53M
19.9 MW
Training power draw
22.6 MW
API access
Accessibility
Open weights (restricted use)
No
Open weights
Yes
United States
Country
United States
Shared benchmarks
Related comparisons
SourceEpoch AI, 'AI Models'. Published online at epoch.ai. Retrieved 2026-07-29 from https://epoch.ai/data/ai-models. Licensed under CC BY 4.0.