#59 Qwen3.6 Flash
medium
Invalid SVG
- Cost
- $0.008
- Time
- 29.9s
- Tokens
- 6,784 tok
Qwen
Release: 2026-04-20
Tested on: 2026-05-21 23:53
qwen/qwen3.6-flash::medium
~35B total (~3B active)MoEClosedEstimated
(medium)
(none)
Summary
Qwen3.6 Flash scores 7.4 on AI BENCHY and ranks #59. It has 9.7 reliability, a 68.3% pass rate, $0.348 total cost, and 18.96s average response time.
What makes Qwen3.6 Flash unique: It stands out most in Instructions following, where it ranks #1, while Coding is its weakest area at #18.
Model facts
Researched on 2026-08-12
- 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 released Flash-class architecture.
7.4
Consistency
8.0
9.7
$0.348
Total Output Tokens
234,489
Total Input Tokens
0
Input Price
$0.188 / 1M
Output Price
$1.125 / 1M
Flaky tests
5
Flaky tests had mixed outcomes across runs (at least one pass and one fail).
Hamster playing table tennis
Prompt: Create a detailed SVG illustration of a hamster playing table tennis.
Run history
| Tested on | Score | Reliability | Tests Correct | Total Cost | Compare |
|---|---|---|---|---|---|
| 2026-08-14 01:43 Re-test | 6.8 | 10.0 | $0.741 ↓ | Compare | |
| 2026-07-16 21:53 New test added | 6.9 | 10.0 | $0.738 ↓ | Compare | |
| 2026-06-04 13:18 New test added | 7.5 | 10.0 | $0.288 ↓ | Compare | |
| 2026-05-21 23:53 Suite changed | 7.4 | 9.7 | $0.348 | Current run | |
| 2026-04-27 23:06 Initial run | 8.1 | 10.0 | $0.174 | Compare |
This run used a different benchmark suite. Keep suite changes in mind when reading historical movement.
Run comparison
| Run | Benchmark coverage | Score | Consistency | Reliability | Tests Correct | Flaky tests | Total Output Tokens | Total Input Tokens | Total Cost | Response Time (avg) |
|---|---|---|---|---|---|---|---|---|---|---|
| 2026-05-21 23:53 · Suite changed | 60/60 attempts | 7.4 | 8.0 | 9.7 | 11/20 | 5 | 234,489 | 0 | $0.348 | 18.96s |
| 2026-08-14 01:43 · Re-test | 66/66 attempts | 6.8 | 8.2 | 10.0 | 12/22 | 5 | 636,949 | 129,050 | $0.741 | 44.97s |
| Difference | — | +0.5 | -0.2 | -0.3 | -1 | 0 | -402460 | -129050 | -$0.394 | -26012ms |
Benchmark coverage differs: 60/60 attempts (Target: 3 repeats per test) versus 66/66 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.
Top Models by Score
Score vs Total Cost
Response Time (avg)
Score vs Response Time (avg)
Total Output Tokens
Score vs Total Output Tokens
Quick Compare
Qwen3.6 FlashmediumvsGemini 3.5 FlashnoneQwen3.6 FlashmediumvsStep 3.7 FlashhighQwen3.6 FlashmediumvsSeed-2.0-CodehighQwen3.6 FlashmediumvsSeed-2.0-CodemediumQwen3.6 FlashmediumvsQwen3.7 FlashmediumQwen3.6 FlashmediumvsSolar Pro 4lowQwen3.6 FlashmediumvsClaude Fable 5.1lowQwen3.6 FlashmediumvsSolar Pro 4mediumQwen3.6 FlashmediumvsKAT-Coder-Pro V2.5mediumQwen3.6 FlashmediumvsDeepSeek V4 Pronone
Category Breakdown
| Category | Score | Consistency | Tests Correct |
|---|---|---|---|
| Anti-AI Tricks | 10.0 | 10.0 | |
| Coding | 5.1 | 6.8 | |
| Combined | 10.0 | 10.0 | |
| Data parsing and extraction | 10.0 | 10.0 | |
| Domain specific | 3.5 | 4.4 | |
| General Intelligence | 4.8 | 9.9 | |
| Instructions following | 10.0 | 10.0 | |
| Puzzle Solving | 6.1 | 4.7 | |
| Tool Calling | 10.0 | 10.0 | |
| Trivia | 3.0 | 10.0 |