Skip to content
NousResearch

Llama 3 70B Instruct

Maker’s page
Check price at Amazon for NousResearch Meta-Llama-3-70B-Instruct (affiliate link, opens in new tab)
Weights at 16-bit
141 GB
Weights at 8-bit (Q8_0)
75 GB
Weights at 4-bit (Q4_K_M)
43 GB
Smallest card at 4-bit
RTX 6000 Ada Generation · 48 GB
Parameters
Layers
KV heads
Context
Licence

Specs

Attention heads
Context
Downloads (30 days)
Head size
Hidden size
KV heads
Layers
Licence
Parameters
Parameters by type
Publisher
Task

More from NousResearch