Alibaba
Qwen3.5 122B A10B (Non-reasoning)
The Qwen3.5 122B-A10B native vision-language model is built on a hybrid architecture that integrates a linear attention mechanism with a sparse mixture-of-experts model, achieving higher inference efficiency. In terms of...
Qwen model releasesModel Snapshot
ReleaseFeb 24, 2026
Context262,144
Overall
Rank #154 across 566
27.6
Coding
Rank #106 across 222
43.3
Blended Price
Rank #188 across 371
$1.10/M
Strength Profile
Percentile score by analysis domain.
* Cost is inverted: lower input, output, and blended prices rank higher.
Benchmark Percentiles
Higher bars mean stronger relative placement.
Model Details
Runtime
Output Speed145.4 tok/s
First Token1.01s
Max Output65,536
Pricing
Blended$1.10/M
Input$0.400/M
All Benchmarks
| Metric | Domain | Value | Rank |
|---|---|---|---|
| Artificial Analysis Intelligence Index | overall | 27.6 | #154 |
| Artificial Analysis Coding Index | coding | 43.3 | #106 |
| GPQA | reasoning | 82.7% | #119 |
| Humanity's Last Exam | reasoning | 14.8% | #142 |
| SciCode |