Llama 3.3 70B

llama

evroc · nvidia/Llama-3.3-70B-Instruct-FP8

Capabilities

Tool CallNo

ReasoningNo

Structured OutputNo

Temperature ControlNo

AttachmentsNo

Open WeightsYes

Input

text

Output

text

Input$1.18

Output$1.18

Direct installation coming soon — March 2026

Providerevroc

Model IDnvidia/Llama-3.3-70B-Instruct-FP8

Familyllama

Release Date2024-12-01

Last Updated2024-12-01

Context131K

Max Output33K

@ai-sdk/openai-compatible