实时
Head to head

Llama 3.1-405B vs U-PaLM (540B)

Meta AI
Llama 3.1-405B
July 2024
vs
Google
U-PaLM (540B)
October 2022
3.8×10²⁵Training compute (FLOP)2.5×10²⁴
$53MTraining cost--

Llama 3.1-405B (Meta AI) and U-PaLM (540B) (Google) are both frontier AI models. Llama 3.1-405B was published in July 2024 and U-PaLM (540B) in October 2022.

Llama 3.1-405B was trained on 3.8×10²⁵ FLOP, about 15.0x the compute of U-PaLM (540B) at 2.5×10²⁴ 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.

These two models share no benchmark on which both have been scored, so no direct performance comparison is possible here. The specification table below is a comparison of inputs, not of results.

Specifications
Meta AI
Organization
Google
Jul 23, 2024
Published
Oct 20, 2022
3.8×10²⁵ FLOP
Training compute15x
2.5×10²⁴ FLOP
405B
Parameters1.3x
540B
15.6T
Dataset size12000x
1.3B
NVIDIA H100 SXM5 80GB
Training hardware
Google TPU v4
16,384
Chips used32x
512
2,142 h
Training time
120 h
$53M
Training cost (2023 USD)
--
22.6 MW
Training power draw
348.3 kW
Open weights (restricted use)
Accessibility
Unreleased
Yes
Open weights
No
United States
Country
United States
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.