Qwen: Qwen3.5-35B-A3B

Qwen3.5-35B-A3B is a multimodal model from Qwen that accepts text, image, and video inputs. It supports tool use and reasoning, making it usable for agentic workflows and multi-step tasks. Its context window of 262,144 tokens accommodates long documents or extended conversations, and it can generate up to 81,920 tokens per response. At $0.14 per million input tokens and $1.00 per million output tokens, it sits on the affordable end for a reasoning-capable multimodal model. Its blended benchmark score of 59.4 is based on only three benchmarks, so treat that figure as a limited signal rather than a comprehensive verdict. The agentic benchmark score of 72.8 is its strongest result, which suggests it may be worth shortlisting for tool-calling and agent-oriented use cases where video and long-context support are also useful. Teams needing broader performance validation should wait for wider benchmark coverage before committing.