AI Hub
← All models

DeepSeek-V4-Flash vs NVIDIA Nemotron 3 Nano 4B

DeepSeek vs NVIDIA — benchmarks, pricing, and capabilities side by side.

  • DeepSeek-V4-Flash has the higher intelligence index (89.4 vs 51.3)
  • NVIDIA Nemotron 3 Nano 4B is cheaper ($0.00 vs $0.10 per 1M input)
DeepSeek-V4-FlashNVIDIA Nemotron 3 Nano 4B
Intelligence index89.451.3
DeveloperDeepSeekNVIDIA
TypeLLMLLM
AccessOpen weights
Context window1,048,576 tokens
Input price$0.10 / 1M$0.00 / 1M
Output price$0.20 / 1M$0.00 / 1M
Speed109 tok/s
ReleasedApril 24, 2026March 16, 2026
Parameters284B (13B active)
Input modalitiesText
Output modalitiesText

Shared benchmarks

DeepSeek-V4-Flash
NVIDIA Nemotron 3 Nano 4B
GPQA Diamond
89.4
51.3
Humanity’s Last Exam
32.1
4.8