Advertise here
#181

Qwen3.5-Flash

Qwen Release: 2026-02-24 Tested on: 2026-10-01 00:59 qwen/qwen3.5-flash-02-23::medium
~35B total (~3B active)MoEClosedEstimated
(medium) (none)

Summary

Qwen3.5-Flash scores 6.8 on AI BENCHY and ranks #181. It has 10.0 reliability, a 66.7% pass rate, $0.184 total cost, and 83.68s average response time.

What makes Qwen3.5-Flash unique: It stands out most in Anti-AI Tricks, where it ranks #1, while Coding is its weakest area at #16.

Model facts

Researched on 2026-08-12

Estimated
Parameters
~35B total (~3B active)
Architecture
MoE
Availability
Closed
License
-

Best estimate from public evidence; the vendor did not disclose every value. Estimate from the nearest released Qwen3.5 Flash-class checkpoint.

Parameter source Availability source

Score

6.8

Consistency

7.9

Reliability

10.0

Total Cost (Current Price)

$0.184

Total Output Tokens

531,668

Total Input Tokens

349,177

Input Price

$0.065 / 1M

Output Price

$0.260 / 1M

Cache Read Price

N/A

Cache Write Price

N/A

Cache prices apply to input tokens. Reads reuse cached prompts; writes store them and can cost extra. Output tokens use the output price.

Tests Correct

Wrong Tests: 11

Attempt pass rate: 66.7%

Flaky tests

6

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

Response Time (avg)

83.68s

Response Time (max): 515.38s

Response Time (total): 1840.92s

Wrong answer: 5 Timed out: 3 API error: 1 Did not follow instructions: 1 Invalid tool call: 1

Hamster playing table tennis

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

#181 Qwen3.5-Flash

medium
Cost
$0.002
Time
25.8s
Tokens
4,294 tok

Run history

Tested on Score Reliability Tests Correct Total Cost Compare
2026-10-01 00:59 Re-test 6.8 10.0 $0.184 Current run
2026-07-16 22:15 New test added 6.2 10.0 $0.139 ↓ Compare
2026-06-04 13:36 New test added 7.4 8.9 $0.080 ↓ Compare
2026-05-21 23:57 Suite changed 7.4 8.3 $0.104 Compare
2026-04-11 01:44 First recorded run 7.8 N/A $0.080 Compare

This run used a different benchmark suite. Keep suite changes in mind when reading historical movement.

Run comparison

RunBenchmark coverageScoreConsistencyReliabilityTests CorrectFlaky testsTotal Output TokensTotal Input TokensTotal CostResponse Time (avg)
2026-10-01 00:59 · Current run69/69 attempts6.87.910.012/236531,668349,177$0.18483.68s
2026-05-21 23:57 · Suite changed60/60 attempts7.47.68.311/206286,0290$0.10465.64s
Difference—-0.6+0.3+1.7+10+245639+349177+$0.081+18042ms

Benchmark coverage differs: 69/69 attempts (Target: 3 repeats per test) versus 60/60 attempts (Target: 3 repeats per test). Totals and repeat-sensitive metrics are not directly comparable.

These two runs used different benchmark suites, so the deltas reflect both model changes and suite changes.

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

Qwen3.5-FlashmediumvsKimi K2.6mediumQwen3.5-FlashmediumvsInklinglowFree AvailableQwen3.5-FlashmediumvsClaude Opus 4.8noneQwen3.5-FlashmediumvsMercury 2.5 PreviewmediumQwen3.5-FlashmediumvsClaude Sonnet 5.5lowQwen3.5-FlashmediumvsGPT-6 SolnoneQwen3.5-FlashmediumvsGPT-5.6 LunalowQwen3.5-FlashmediumvsSpace Bunny AlphamaxQwen3.5-FlashmediumvsGemma 4 31BmediumFree AvailableQwen3.5-FlashmediumvsSolar Pro 4xhigh

Category Breakdown

Category Score Consistency Tests Correct
Agentic 10.0 10.0
Anti-AI Tricks 10.0 10.0
Coding 3.7 7.2
Combined 6.4 5.8
Data parsing and extraction 7.3 5.9
Domain specific 2.9 7.2
General Intelligence 6.1 3.1
Instructions following 10.0 10.0
Puzzle Solving 8.2 7.2
Tool Calling 10.0 10.0
Trivia 3.0 10.0

Compared models