#102 Qwen3.5-35B-A3B
none- Cost
- $0.005
- Time
- 28.4s
- Tokens
- 4,518 tok
Qwen
Release: 2026-02-24
Tested on: 2026-05-22 00:00
qwen/qwen3.5-35b-a3b::none
35B total (3B active)MoEOpen source
(medium)
(none)
Summary
Qwen3.5-35B-A3B scores 5.8 on AI BENCHY and ranks #102. It has 10.0 reliability, a 45.0% pass rate, $0.016 total cost, and 3.50s average response time.
What makes Qwen3.5-35B-A3B unique: It stands out most in Domain specific, where it ranks #2, while Anti-AI Tricks is its weakest area at #18. It is notably fast compared with similar models.
Model facts
Researched on 2026-08-12
- Parameters
- 35B total (3B active)
- Architecture
- MoE
- Availability
- Open source
- License
- Apache-2.0
5.8
Consistency
8.9
10.0
$0.016
Total Output Tokens
4,334
Total Input Tokens
0
Input Price
$0.139 / 1M
Output Price
$1.000 / 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.
Flaky tests
3
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 01:53 Re-test | 5.9 | 10.0 | $0.187 ↑ | Compare | |
| 2026-07-16 22:11 New test added | 6.1 | 10.0 | $0.106 ↓ | Compare | |
| 2026-06-04 13:29 New test added | 5.6 | 10.0 | $0.012 ↓ | Compare | |
| 2026-05-22 00:00 Suite changed | 5.8 | 10.0 | $0.016 | Current run | |
| 2026-04-11 01:19 First recorded run | 6.1 | N/A | $0.016 | 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) |
|---|---|---|---|---|---|---|---|---|---|---|
| 2026-05-22 00:00 · Suite changed | 60/60 attempts | 5.8 | 8.9 | 10.0 | 7/20 | 3 | 4,334 | 0 | $0.016 | 3.50s |
| 2026-08-14 01:53 · Re-test | 66/66 attempts | 5.9 | 8.6 | 10.0 | 6/22 | 4 | 86,616 | 134,530 | $0.187 | 12.73s |
| Difference | — | -0.1 | +0.3 | 0.0 | +1 | -1 | -82282 | -134530 | -$0.171 | -9226ms |
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
Qwen3.5-35B-A3BnonevsGLM 5noneQwen3.5-35B-A3BnonevsNemotron 3 SupermediumFree AvailableQwen3.5-35B-A3BnonevsStep 5 PreviewlowQwen3.5-35B-A3BnonevsKimi K2.6noneQwen3.5-35B-A3BnonevsSolar Pro 4noneQwen3.5-35B-A3BnonevsLing 3.1 FlashnoneQwen3.5-35B-A3BnonevsGPT-5.4noneQwen3.5-35B-A3BnonevsGemini 2.5 FlashnoneQwen3.5-35B-A3BnonevsSpace Bunny AlphalowQwen3.5-35B-A3BnonevsGPT-6 Lunanone
Category Breakdown
| Category | Score | Consistency | Tests Correct |
|---|---|---|---|
| Anti-AI Tricks | 3.4 | 7.9 | |
| Coding | 6.8 | 10.0 | |
| Combined | 3.0 | 10.0 | |
| Data parsing and extraction | 10.0 | 10.0 | |
| Domain specific | 7.7 | 10.0 | |
| General Intelligence | 6.5 | 3.4 | |
| Instructions following | 6.3 | 10.0 | |
| Puzzle Solving | 3.7 | 7.4 | |
| Tool Calling | 10.0 | 10.0 | |
| Trivia | 3.0 | 10.0 |