Zaynoid/Qwen3.5-9B-EBOS-v1

VISIONConcurrency Cost:1Model Size:9BQuant:FP8Ctx Length:32kTool Calling:SupportedPublished:Apr 25, 2026License:apache-2.0Architecture:Transformer Open Weights Cold

Qwen3.5-9B-EBOS-v1 is a 9 billion parameter multimodal large language model developed by Qwen, featuring a unified vision-language foundation and an efficient hybrid architecture. It excels in multimodal reasoning, coding, and agentic tasks, supporting a native context length of 262,144 tokens. The model integrates Gated Delta Networks and sparse Mixture-of-Experts for high-throughput inference and is optimized for robust real-world adaptability through scalable reinforcement learning across 201 languages.

Loading preview...

Overview

Qwen3.5-9B is a 9 billion parameter multimodal large language model from Qwen, designed for exceptional utility and performance. It features a unified vision-language foundation, achieving cross-generational parity with Qwen3 and outperforming Qwen3-VL models across various benchmarks including reasoning, coding, agents, and visual understanding. The model incorporates an efficient hybrid architecture using Gated Delta Networks and sparse Mixture-of-Experts for high-throughput inference with minimal latency.

Key Capabilities

  • Multimodal Learning: Early fusion training on multimodal tokens enables strong performance in both language and vision tasks.
  • Scalable RL Generalization: Enhanced real-world adaptability through reinforcement learning scaled across million-agent environments.
  • Global Linguistic Coverage: Supports 201 languages and dialects for inclusive, worldwide deployment.
  • Extended Context Length: Natively handles up to 262,144 tokens, extensible to 1,010,000 tokens with RoPE scaling techniques like YaRN.
  • Agentic Usage: Excels in tool calling, with recommended integration via Qwen-Agent and Qwen Code.

What Makes It Different

Qwen3.5-9B stands out due to its unified vision-language foundation that allows it to perform comparably to larger, specialized models in multimodal tasks. Its hybrid architecture ensures efficient inference, making it suitable for demanding applications. The model also defaults to a "thinking mode" for complex tasks, generating internal reasoning before final responses, which can be configured for direct output. Benchmarks show strong performance in knowledge, instruction following, long context, reasoning, coding, and general agent tasks, often surpassing previous Qwen3 models and competitive with much larger models.

Finetunes
2 models