Skip to content

Qwen3.5 9B

Qwen/Qwen3.5-9B
Open Source · chat · open-weights
GA Alert me on changes
Intelligence
Context
262.1K
Max output
Input $/1M
$0.10
Output $/1M
$0.15
Modalities
text · image
Released
27 Feb 2026
Intelligence Index via Artificial Analysis · 0–100, higher is better
License: apache-2.0 · Qwen/Qwen3.5-9B
AI summary
● machine-written

Qwen3.5-9B: multimodal 9B model with text and image support

Qwen3.5-9B is a 9-billion-parameter open-weights multimodal model from Alibaba's Qwen family, supporting both text and image inputs. The model features a unified vision-language architecture with early fusion of multimodal tokens, enabling reasoning across text and images in a single context. It is available under the Apache 2.0 license and released on March 10, 2026.

What's new
  • Unified vision-language design with early fusion of multimodal tokens
  • 262K token context window
  • Pricing at $0.10 per million input tokens, $0.15 per million output tokens
  • Available from 7 API providers including Parasail, Venice, and SiliconFlow
Best for
Visual understanding tasks requiring image and text reasoningCoding and reasoning applicationsCost-efficient multimodal inference at 9B parameters
Sources

Source: https://huggingface.co/Qwen/Qwen3.5-9B