Qwen3.5 397B-A17B is the largest and first-released model in Alibaba's Qwen3.5 mixture-of-experts generation, targeting flagship-level reasoning, coding, and agentic performance among open-weight models.
Cost rate
0.6x
Context
262K
Released
Feb 16, 2026
Input
TextImageVideo
Output
Text
Support
ReasoningTool callingStructured outputs
Performance
Median latency and throughput measured across recent requests.
Throughput46 tok/s
Latency2.47s
Best at
The categories where this model ranks highest.
Image understanding
#59 路 top 35%
Business, Management, & Finance
#87 路 top 21%
Medicine & Healthcare
#87 路 top 22%
Life, Physical, & Social Science
#88 路 top 21%
| Domain | Score | Rank |
|---|---|---|
| CodingCode Arena Elo | 1400 | #108 |
| Software & IT Services | 1480 | #100 |
| Writing, Literature, & Language | 1421 | #98 |
| Life, Physical, & Social Science | 1465 | #88 |
| Entertainment, Sports, & Media | 1403 | #108 |
| Business, Management, & Finance | 1447 | #87 |
| Mathematical | 1453 | #88 |
| Legal & Government | 1447 | #103 |
| Medicine & Healthcare | 1467 | #87 |
| Multimodal | ||
| Image understanding | 1246 | #59 |
| Category | Win rate | Rank |
|---|---|---|
| SVG | 56.1% | #60 |
| Data Visualization | 53.2% | #72 |
| Code Categories | 52.7% | #75 |
| Website | 52.6% | #77 |
| 3D | 56.7% | #78 |
| UI Component | 51.4% | #85 |
| Game Development | 50.2% | #88 |
Latest news about Qwen3.5 397B A17B
Recent articles and community posts, linked to their original sources.
Qwenputer: A functioning, deterministic RISC-V CPU inside Qwen3.5Hacker News709147- Kev: Tiny Jev-like family of decision models built on top of Qwen3.5Hacker News463210
Mapping every neuron in Qwen3-8BHacker News302
Doubling Qwen Flash Next speed with Strata HotRodNo place like localhost18,864
Qwen 4 Max is here! OpenAI and Anthropic are in troubleSimon H酶iberg17,830
Qwen4 Max Just Dropped, China's New BEST AI Model!AI Revolution84,709