SmophyAI

Published daily · Next update 02:00 UTC

Model family · 39 versions tracked

Qwen: every version compared

Qwen: Qwen3.8 Max currently leads the Qwen family with an Artificial Analysis Intelligence Index of 53.4. For the best value within the family, Qwen: Qwen3.5-9B costs $0.19/1M tokens blended.

  1. Qwen: Qwen3.8 Max - 53.4 intelligence · $3.00/1M
  2. Qwen: Qwen3.8 2.4T A95B - 39.9 intelligence · $3.00/1M
  3. Qwen: Qwen3.7 Max - 29.5 intelligence · $2.21/1M
  4. Qwen: Qwen3.6 Max Preview - 28.4 intelligence · $2.31/1M
  5. Qwen: Qwen3.7 Plus (batch) - 25.2 intelligence · $0.56/1M
  6. Qwen: Qwen3.5-35B-A3B - 24.3 intelligence · $0.35/1M
  7. Qwen: Qwen3.5-27B - 22.9 intelligence · $0.84/1M
  8. Qwen: Qwen3.6 27B - 21.4 intelligence · $1.01/1M
  9. Qwen: Qwen3.5 397B A17B - 21.4 intelligence · $1.28/1M
  10. Qwen: Qwen3 Max Thinking - 21.3 intelligence · $1.56/1M
  11. Qwen: Qwen3.8 27B (batch) - 20.1 intelligence · $0.58/1M
  12. Qwen: Qwen3.6 35B A3B - 18.2 intelligence · $0.30/1M
  13. Qwen: Qwen3.5-122B-A10B - 15.6 intelligence · $0.71/1M
  14. Qwen: Qwen3 235B A22B Thinking 2507 - 12.7 intelligence · $1.21/1M
  15. Qwen: Qwen3.5-9B - 11.2 intelligence · $0.19/1M
  16. Qwen: Qwen3 30B A3B Thinking 2507 - 9.8 intelligence · $0.75/1M
  17. Qwen: Qwen3 Coder Next - 9.2 intelligence · $0.29/1M
  18. Qwen: Qwen3.7 Flash - - intelligence · $0.05/1M
  19. Qwen: Qwen3.6 Flash - - intelligence · $0.42/1M
  20. Qwen: Qwen-Plus - - intelligence · $0.39/1M
  21. Qwen: Qwen3 VL 32B Instruct - - intelligence · $0.18/1M
  22. Qwen: Qwen3 Coder Flash - - intelligence · $0.39/1M
  23. Qwen: Qwen3 VL 30B A3B Thinking - - intelligence · $0.47/1M
  24. Qwen: Qwen3 VL 30B A3B Instruct - - intelligence · $0.26/1M
  25. Qwen: Qwen3 VL 8B Thinking - - intelligence · $0.66/1M
  26. Qwen: Qwen3 VL 8B Instruct - - intelligence · $0.38/1M
  27. Qwen: Qwen3 VL 235B A22B Thinking - - intelligence · $1.30/1M
  28. Qwen: Qwen3 Max - - intelligence · $1.56/1M
  29. Qwen: Qwen3 Coder Plus - - intelligence · $1.30/1M
  30. Qwen: Qwen3 VL 235B A22B Instruct - - intelligence · $0.63/1M
  31. Qwen: Qwen3 30B A3B Instruct 2507 - - intelligence · $0.15/1M
  32. Qwen: Qwen2.5 7B Instruct - - intelligence · $0.13/1M
  33. Qwen2.5 72B Instruct - - intelligence · $0.37/1M
  34. Qwen: Qwen3 Coder 30B A3B Instruct - - intelligence · $0.26/1M
  35. Qwen: Qwen2.5 VL 72B Instruct - - intelligence · $0.85/1M
  36. Qwen2.5 Coder 32B Instruct - - intelligence · $0.75/1M
  37. Qwen: Qwen3.8 Flash - - intelligence · $0.23/1M
  38. Qwen: Qwen3.8 Omni Flash - - intelligence · $0.23/1M
  39. Qwen: Qwen3.8 Max Prime - - intelligence · $6.00/1M
All 39 Qwen versions, by quality
#ModelIntelligence indexPrice/1M blended
01Qwen: Qwen3.8 Max53.4$3.00
02Qwen: Qwen3.8 2.4T A95B39.9$3.00
03Qwen: Qwen3.7 Max29.5$2.21
04Qwen: Qwen3.6 Max Preview28.4$2.31
05Qwen: Qwen3.7 Plus (batch)25.2$0.56
06Qwen: Qwen3.5-35B-A3B24.3$0.35
07Qwen: Qwen3.5-27B22.9$0.84
08Qwen: Qwen3.6 27B21.4$1.01
09Qwen: Qwen3.5 397B A17B21.4$1.28
10Qwen: Qwen3 Max Thinking21.3$1.56
11Qwen: Qwen3.8 27B (batch)20.1$0.58
12Qwen: Qwen3.6 35B A3B18.2$0.30
13Qwen: Qwen3.5-122B-A10B15.6$0.71
14Qwen: Qwen3 235B A22B Thinking 250712.7$1.21
15Qwen: Qwen3.5-9B11.2$0.19
16Qwen: Qwen3 30B A3B Thinking 25079.8$0.75
17Qwen: Qwen3 Coder Next9.2$0.29
18Qwen: Qwen3.7 Flash-$0.05
19Qwen: Qwen3.6 Flash-$0.42
20Qwen: Qwen-Plus-$0.39
21Qwen: Qwen3 VL 32B Instruct-$0.18
22Qwen: Qwen3 Coder Flash-$0.39
23Qwen: Qwen3 VL 30B A3B Thinking-$0.47
24Qwen: Qwen3 VL 30B A3B Instruct-$0.26
25Qwen: Qwen3 VL 8B Thinking-$0.66
26Qwen: Qwen3 VL 8B Instruct-$0.38
27Qwen: Qwen3 VL 235B A22B Thinking-$1.30
28Qwen: Qwen3 Max-$1.56
29Qwen: Qwen3 Coder Plus-$1.30
30Qwen: Qwen3 VL 235B A22B Instruct-$0.63
31Qwen: Qwen3 30B A3B Instruct 2507-$0.15
32Qwen: Qwen2.5 7B Instruct-$0.13
33Qwen2.5 72B Instruct-$0.37
34Qwen: Qwen3 Coder 30B A3B Instruct-$0.26
35Qwen: Qwen2.5 VL 72B Instruct-$0.85
36Qwen2.5 Coder 32B Instruct-$0.75
37Qwen: Qwen3.8 Flash-$0.23
38Qwen: Qwen3.8 Omni Flash-$0.23
39Qwen: Qwen3.8 Max Prime-$6.00

Qwen real usage share, over time

2.6%1.9%1.3%0.6%0.0%Share of usageJul 5Sep 7Nov 6

Ranked by current quality, not release date - this pipeline's data collection started recently, so "first tracked" timestamps don't reflect true historical release dates yet. The usage-share chart above is real, unaffected by that limitation.

Frequently asked questions

What is the best Qwen model?

Qwen: Qwen3.8 Max currently leads the Qwen family with an Artificial Analysis Intelligence Index of 53.4.

How much real usage does Qwen get?

Qwen models account for 1.0% of real OpenRouter token volume as of the latest data.