GeekRadar / The comparison desk

COMPARE.

AI models

meta

Llama 3-8B

AI model Released Apr 18, 2024

Open weights: Yes^*

Capability & preference

General capability · ECI

A cross-benchmark capability estimate. Index points, not a percentage; intervals show uncertainty.

Llama 3-8B 109.6^* Llama 3-8B · 95% interval 104.1–114.4

Run it locally

Inference · memory estimate

Llama 3-8B

Not enough verified architecture information for a reliable VRAM estimate.

Capabilities

Open weights: Yes^*

System configuration

Detailed specifications

Specifications · 4 fields

Access model: Open weights (restricted use)^*
Open weights: Yes^*
Organization: Meta AI^*
Organization country: United States of America^*

* Provisional data