AI Hub
← All models

DeepSeek-V4-Flash vs Granite 4.0 H 350M

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

  • DeepSeek-V4-Flash has the higher intelligence index (89.4 vs 10.4)
  • Granite 4.0 H 350M is cheaper ($0.00 vs $0.10 per 1M input)
DeepSeek-V4-FlashGranite 4.0 H 350M
Intelligence index89.410.4
DeveloperDeepSeekIBM
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, 2026October 28, 2025
Parameters284B (13B active)
Input modalitiesText
Output modalitiesText

Shared benchmarks

DeepSeek-V4-Flash
Granite 4.0 H 350M
GPQA Diamond
89.4
25.7
Humanity’s Last Exam
32.1
6.4