AI Hub
← All models

Gemini 3 Flash vs Qwen3 VL 235B A22B Instruct

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

  • •Gemini 3 Flash has the higher intelligence index (90.2 vs 70.9)
  • •Qwen3 VL 235B A22B Instruct is cheaper ($0.20 vs $0.50 per 1M input)
  • •Gemini 3 Flash is faster
  • •Gemini 3 Flash has a larger context window (1M)
Gemini 3 FlashQwen3 VL 235B A22B Instruct
Intelligence index90.270.9
DeveloperGoogleAlibaba
TypeMultimodalMultimodal
AccessAPI onlyOpen weights
Context window1,048,576 tokens262,144 tokens
Input price$0.50 / 1M$0.20 / 1M
Output price$3.00 / 1M$0.88 / 1M
Speed191 tok/s51 tok/s
ReleasedDecember 17, 2025September 23, 2025
Parameters——
Input modalitiesText, Image, Audio, VideoText, Image
Output modalitiesTextText

Shared benchmarks

Gemini 3 Flash
Qwen3 VL 235B A22B Instruct
AIME 2025
97
70.7
GPQA Diamond
90.4
71.2
Humanity’s Last Exam
34.7
6.3
LiveCodeBench
90.8
59.4
MMLU-Pro
89
82.3
Gemini 3 Flash details Qwen3 VL 235B A22B Instruct details