AI Hub
← All models

Granite 4.0 H Small vs Qwen3.6 Max

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

  • Qwen3.6 Max has the higher intelligence index (88.8 vs 35.7)
  • Granite 4.0 H Small is cheaper ($0.10 vs $1.04 per 1M input)
  • Granite 4.0 H Small is faster
Granite 4.0 H SmallQwen3.6 Max
Intelligence index35.788.8
DeveloperIBMAlibaba
TypeLLMLLM
AccessAPI only
Context window262,144 tokens
Input price$0.10 / 1M$1.04 / 1M
Output price$0.30 / 1M$6.24 / 1M
Speed524 tok/s36 tok/s
ReleasedSeptember 22, 2025April 27, 2026
Parameters
Input modalitiesText
Output modalitiesText

Shared benchmarks

Granite 4.0 H Small
Qwen3.6 Max
GPQA Diamond
41.6
88.8
Humanity’s Last Exam
3.7
28.9