Compare AI models

Specifications
Parameters
236B8B/49B
Context window
128k128k
These models have no shared benchmark scores.
Overview
CompanyDeepSeekNVIDIA
Release dateJun 28 2024Mar 18 2025
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

DeepSeek V2-0628 and Llama Nemotron (Nano/Super) don't publish scores on any of the same benchmarks, so there's no direct head-to-head comparison. DeepSeek V2-0628 shipped 263 days before Llama Nemotron (Nano/Super), so benchmark comparisons should account for the intervening progress.

DeepSeek V2-0628 has 236B parameters, while Llama Nemotron (Nano/Super) has 8B/49B. Context windows are 128k (DeepSeek V2-0628) vs 128k (Llama Nemotron (Nano/Super)).

Direct benchmark comparisons are unavailable — DeepSeek V2-0628 and Llama Nemotron (Nano/Super) don't publish scores on any of the same benchmarks.