AI Hub
← All models

DeepSeek-V4-Flash vs Qwen2 7B Instruct

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

  • DeepSeek-V4-Flash has the higher intelligence index (89.4 vs 37.4)
DeepSeek-V4-FlashQwen2 7B Instruct
Intelligence index89.437.4
DeveloperDeepSeekAlibaba
TypeLLMLLM
AccessOpen weightsOpen weights
Context window1,048,576 tokens
Input price$0.10 / 1M
Output price$0.20 / 1M
Speed109 tok/s
ReleasedApril 24, 2026July 23, 2024
Parameters284B (13B active)7620000000
Input modalitiesText
Output modalitiesText

Shared benchmarks

DeepSeek-V4-Flash
Qwen2 7B Instruct
GPQA Diamond
89.4
25.3