qwen/qwen3.5-35b-a3bQwen3.5-35B-A3B is Alibaba’s open-weights AI model, released Feb 24, 2026. It costs $0.16/M input and $1.30/M output tokens, and handles a 262K-token context window.
The Qwen3.5 Series 35B-A3B is a native vision-language model designed with a hybrid architecture that integrates linear attention mechanisms and a sparse mixture-of-experts model, achieving higher inference efficiency.
Scores on standardized evaluations. Higher is better — and rank shows where Qwen3.5-35B-A3B lands among all models tracked on Model Beat.
Graduate-level scientific reasoning
Frontier of human expert knowledge
Scientific research coding
Olympiad-qualifier math
Tool-agent-user reliability
The Qwen3.5 Series 35B-A3B is a native vision-language model designed with a hybrid architecture that integrates linear attention mechanisms and a sparse mixture-of-experts model, achieving higher inference efficiency.
Qwen3.5-35B-A3B is an AI model developed by Alibaba, part of the Qwen family, released Feb 24, 2026.
Qwen3.5-35B-A3B's list price from Alibaba is $0.16 per million input tokens and $1.30 per million output tokens. The cheapest credible third-party provider on OpenRouter (DeepInfra) serves it at $0.14/$1 per 1M.
Qwen3.5-35B-A3B supports a context window of up to 262K tokens.
On standardized evaluations tracked by Epoch AI, Qwen3.5-35B-A3B scores 89.2% on τ²-bench, 83.5% on GPQA Diamond, 21.0% on Humanity's Last Exam.
Qwen3.5-35B-A3B is released with open weights (Open weights (unrestricted)), so the model can be downloaded and self-hosted.
Benchmark data from Epoch AI (CC BY) and Artificial Analysis; pricing, specs & descriptions from OpenRouter. Data updated Sep 10, 2026.