Compare AI models

Specifications
Parameters
284B49B
Context window
1M128k
API pricing
Cheapest input
$0.0383Relace—
Cheapest output
$0.0806Baidu—
$12.09Baidu · fp8—
Benchmarks
LiveCodeBench
91.6%73.58%
GPQA Diamond
88.1%71.97%
Overview
CompanyDeepSeekNVIDIA
Release dateApr 24 2026Jul 25 2025
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

DeepSeek-V4-Flash leads Llama Nemotron Super 49B v1.5 on 2 of the 2 benchmarks they both report (LiveCodeBench, GPQA Diamond). Llama Nemotron Super 49B v1.5 shipped 273 days before DeepSeek-V4-Flash, so benchmark comparisons should account for the intervening progress.

DeepSeek-V4-Flash has 284B parameters, while Llama Nemotron Super 49B v1.5 has 49B. Context windows are 1M (DeepSeek-V4-Flash) vs 128k (Llama Nemotron Super 49B v1.5).

On LiveCodeBench, DeepSeek-V4-Flash leads at 91.6% vs Llama Nemotron Super 49B v1.5 at 73.58%. On GPQA Diamond, DeepSeek-V4-Flash leads at 88.1% vs Llama Nemotron Super 49B v1.5 at 71.97%.