AI Hub
← All models

Qwen3 4B 2507 Instruct vs Qwen3.6 Max

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

  • Qwen3.6 Max has the higher intelligence index (88.8 vs 52.2)
  • Qwen3 4B 2507 Instruct is cheaper ($0.00 vs $1.04 per 1M input)
Qwen3 4B 2507 InstructQwen3.6 Max
Intelligence index52.288.8
DeveloperAlibabaAlibaba
TypeLLMLLM
AccessAPI only
Context window262,144 tokens
Input price$0.00 / 1M$1.04 / 1M
Output price$0.00 / 1M$6.24 / 1M
Speed36 tok/s
ReleasedAugust 6, 2025April 27, 2026
Parameters
Input modalitiesText
Output modalitiesText

Shared benchmarks

Qwen3 4B 2507 Instruct
Qwen3.6 Max
GPQA Diamond
51.7
88.8
Humanity’s Last Exam
4.7
28.9