AI BENCHY
Advertise here
#23

Qwen3.8 Max (0902)

Qwen Release: 2026-09-07 Tested on: 2026-09-07 16:06 qwen/qwen3.8-max-0902::low
2.4T total (~100B active)MoEClosedEstimated
(high) (medium) (low)

Summary

Qwen3.8 Max (0902) scores 9.1 on AI BENCHY and ranks #23. It has 10.0 reliability, a 86.4% pass rate, $0.664 total cost, and 24.21s average response time.

Model facts

Researched on 2026-09-07

Estimated
Parameters
2.4T total (~100B active)
Architecture
MoE
Availability
Closed
License
-

Best estimate from public evidence; the vendor did not disclose every value. OpenRouter reports 2400B total and MoE for this snapshot. Active count is undisclosed; 100B is a low-confidence estimate carried from the preceding Qwen3.8 Max catalog record, not a reported snapshot specification. QwenCloud identifies this as an API snapshot.

Parameter source 1 Parameter source 2 Availability source

Score

9.1

Consistency

9.3

Reliability

10.0

Total Cost (Current Price)

$0.664

Total Output Tokens

75,865

Total Input Tokens

103,954

Input Price

$2.000 / 1M

Output Price

$6.000 / 1M

Tests Correct

Wrong Tests: 4

Attempt pass rate: 86.4%

Flaky tests

2

Flaky tests had mixed outcomes across runs (at least one pass and one fail).

Response Time (avg)

24.21s

Response Time (max): 207.79s

Response Time (total): 532.62s

Wrong answer: 4

Hamster playing table tennis

Prompt: Create a detailed SVG illustration of a hamster playing table tennis.

#23 Qwen3.8 Max (0902)

low
Cost
$0.014
Time
41.2s
Tokens
2,421 tok

Charts

Choose the first model, then click a second model to open a side-by-side page.

Total Output Tokens

Score vs Total Output Tokens

Quick Compare

Qwen3.8 Max (0902)lowvsGemini 3.8 FlashmediumQwen3.8 Max (0902)lowvsSeed 2.1 TurbolowQwen3.8 Max (0902)lowvsClaude Opus 5mediumQwen3.8 Max (0902)lowvsQwen3.7 MaxmediumQwen3.8 Max (0902)lowvsClaude Opus 5highQwen3.8 Max (0902)lowvsClaude Fable 5.1highQwen3.8 Max (0902)lowvsGemini 3.7 FlashlowQwen3.8 Max (0902)lowvsGemini 3.5 FlashmediumQwen3.8 Max (0902)lowvsGemini 3.1 Pro PreviewmediumQwen3.8 Max (0902)lowvsGPT-5.5medium

Category Breakdown

Category Score Consistency Tests Correct
Anti-AI Tricks 10.0 10.0
Coding 10.0 10.0
Combined 10.0 10.0
Data parsing and extraction 10.0 10.0
Domain specific 5.3 7.2
General Intelligence 6.1 3.1
Instructions following 10.0 10.0
Puzzle Solving 10.0 10.0
Tool Calling 10.0 10.0
Trivia 3.0 10.0

Compared models