Ao vivo
Head to head

Llama 4 Behemoth (preview) vs PaLM 2

vs
Google
PaLM 2
May 2023
5.2×10²⁵Training compute (FLOP)7.3×10²⁴
$45MTraining cost$5M

Llama 4 Behemoth (preview) (Meta AI) and PaLM 2 (Google) are both frontier AI models. Llama 4 Behemoth (preview) was published in April 2025 and PaLM 2 in May 2023.

Llama 4 Behemoth (preview) was trained on 5.2×10²⁵ FLOP, about 7.1x the compute of PaLM 2 at 7.3×10²⁴ 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
Apr 5, 2025
Published
May 10, 2023
5.2×10²⁵ FLOP
Training compute7.1x
7.3×10²⁴ FLOP
2T
Parameters5.9x
340B
30T
Dataset size8.3x
3.6T
NVIDIA H100 SXM5 80GB
Training hardware
Google TPU v4
32,000
Chips used
--
$45M
Training cost (2023 USD)8.9x
$5M
43.9 MW
Training power draw
--
Unreleased
Accessibility
API access
No
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.