Llama-3.1-Nemotron-Ultra-253B-v1

Nebius Token Factory · nvidia/Llama-3_1-Nemotron-Ultra-253B-v1

Capabilities

Tool CallYes

ReasoningNo

Structured OutputYes

Temperature ControlYes

AttachmentsNo

Open WeightsYes

Input

text

Output

text

Input$0.60

Output$1.80

Cache Read$0.06

Cache Write$0.75

Direct installation coming soon — March 2026

ProviderNebius Token Factory

Model IDnvidia/Llama-3_1-Nemotron-Ultra-253B-v1

Knowledge Cutoff2024-12

Release Date2025-01-15

Last Updated2026-02-04

Context128K

Max Input120K

Max Output4K

@ai-sdk/openai-compatible