Compare AI models

Specifications
Parameters
671B253B
Context window
128k128k
API pricing
Cheapest input
$0.256DeepInfra—
Cheapest output
$0.712DeepInfra—
$103StreamLake—
Benchmarks
LiveCodeBench
40.5%66.3%
GPQA Diamond
59.1%76%
Overview
CompanyDeepSeekNVIDIA
Release dateDec 26 2024Apr 7 2025
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

Llama Nemotron Ultra 253B leads DeepSeek V3 on 2 of the 2 benchmarks they both report (LiveCodeBench, GPQA Diamond). DeepSeek V3 shipped 102 days before Llama Nemotron Ultra 253B, so benchmark comparisons should account for the intervening progress.

DeepSeek V3 has 671B parameters, while Llama Nemotron Ultra 253B has 253B. Context windows are 128k (DeepSeek V3) vs 128k (Llama Nemotron Ultra 253B).

On LiveCodeBench, Llama Nemotron Ultra 253B leads at 66.3% vs DeepSeek V3 at 40.5%. On GPQA Diamond, Llama Nemotron Ultra 253B leads at 76% vs DeepSeek V3 at 59.1%.