Qwen3.6 35B A3B
Qwen/Qwen3.6-35B-A3B
Open Source · chat · open-weights
Intelligence
Context
262.1K
Max output
—
Input $/1M
$0.14
Output $/1M
$0.90
Modalities
text · image
Released
15 Apr 2026
Intelligence Index via Artificial Analysis · 0–100, higher is better
License: apache-2.0 · Qwen/Qwen3.6-35B-A3B
AI summary
● machine-written
Qwen3.6 35B A3B: Open-weight multimodal model with 262K context
Qwen3.6 35B A3B is an open-weight multimodal model from Alibaba Cloud with 35 billion total parameters and 3 billion active parameters per token, featuring a hybrid sparse mixture-of-experts architecture. The model supports text, image, and video inputs with a 262K token native context window and integrated thinking mode with reasoning traces. It is available across 7 API providers starting at $0.14 per million input tokens and $0.90 per million output tokens.
What's new
- Hybrid sparse mixture-of-experts with Gated DeltaNet linear attention and gated attention layers
- 262K native context window (extensible to 1M via YaRN)
- Integrated thinking mode with reasoning traces preserved across multi-turn conversations
- Supports function calling and structured output
- Released April 27, 2026
Best for
Multimodal tasks requiring text, image, and video processingApplications requiring extended context windowsCost-efficient inference at scale