TellWorldAI World Archive
Ask中文
News
Models
Organisations
People
Papers & concepts
Products & agents
Chips & robots
History
Topics
Ask中文

Home›Models

Model

Qwen3.5

Qwen3.5 is a natively multimodal model family from Alibaba released in February 2026. Its open flagship, Qwen3.5-397B-A17B, has 397B total parameters and 17B active per token, under Apache 2.0.

Status AvailableModality Text, Image, VideoOpen weights YesContext (K tokens) 256 Era Generative AI for everyone

Parameters
397B
License
Apache-2.0
Access
API, App, Download
Release date
Developer
Alibaba Cloud

Background

Qwen3.5 is the generation of Alibaba's Qwen models released in February 2026. Its flagship open model, Qwen3.5-397B-A17B, was posted to Hugging Face on 16 February 2026. Smaller open models followed in the last days of that month. [2][3]

What it is

Qwen3.5 is natively multimodal. It is a unified vision-language foundation trained with early fusion on multimodal tokens. Early fusion means that text and visual inputs are combined into one stream of tokens during training, instead of being handled by separate encoders that are merged later. [1]

The flagship has 397 billion total parameters, with 17 billion activated for each token. Its 60 layers combine Gated DeltaNet and Gated Attention blocks. The mixture-of-experts part uses 512 experts, with 10 routed and 1 shared expert active per token. [1]

Model sizes

Between 24 and 28 February 2026, Alibaba posted smaller open models: 122B-A10B, 35B-A3B, 27B, 9B, 4B, 2B and 0.8B. The naming follows the same convention as the flagship, in which the number before the letter gives total size and the A-number gives the activated size. [2]

Context and languages

The native context window is 262,144 tokens, which can be extended to about one million tokens. Language support was expanded to 201 languages and dialects. [1]

Benchmarks

On language tasks, Alibaba reported MMLU-Pro 87.8, GPQA 88.4, LiveCodeBench v6 83.6, AIME26 91.3, SWE-bench Verified 76.4, Terminal Bench 2 at 52.5 and BrowseComp 69.0. On vision and agent tasks it reported MMMU 85.0, OmniDocBench1.5 90.8, OSWorld-Verified 62.2 and VideoMME with subtitles 87.5. These scores are the developer's own reports and measure specific tests, so they show capability in those areas rather than overall quality. [1]

Hosted version and agents

The hosted Qwen3.5-Plus, offered on Alibaba Cloud Model Studio, has a 1-million-token default context, built-in tools and adaptive tool use, meaning the model decides when to call a tool. Alibaba said Qwen3.5 can operate desktop and mobile applications. [1][3]

Related releases

Qwen3-Coder-Next was released in the same month. [3]

Position in the family

Later releases in the family include Qwen3.6, which came in April 2026, and Qwen3.8, which came in August 2026. Readers comparing these generations should check which models are open and which are available only as hosted services. [1][2]

CapabilitiesLarge language modelsMultimodalAgents & tool use

Ecosystem & regionOpen sourceChina

Version history

14
  1. 2023
    • Aug 3, 2023 Qwen (Tongyi Qianwen, 2023)

      Qwen, known in Chinese as Tongyi Qianwen (通义千问), is Alibaba's family of large language models. Its first open-weight model, Qwen-7B, was released on 3 August 2023, and the family has grown through many later versions.

  2. 2024
    • Feb 5, 2024 Qwen1.5

      Qwen1.5 is the Qwen family's open model release of February 2024, with base and chat models from 0.5B to 72B parameters, later extended to 110B, and a mixture-of-experts model.

    • Jun 7, 2024 Qwen2

      Qwen2 is the Qwen family's June 2024 generation, with five sizes from 0.5B to 72B parameters, context lengths up to 128K tokens, and most sizes released under Apache 2.0.

    • Sep 19, 2024 Qwen2.5

      Qwen2.5 is the Qwen family's September 2024 generation. Its launch included 13 open-weight models, with up to 18 trillion pretraining tokens, 128K context and most models under Apache 2.0.

    • Nov 28, 2024 QwQ-32B-Preview

      QwQ-32B-Preview is an experimental reasoning research model released by the Qwen team on 28 November 2024, with 32B parameters, a 32K-token context and open Apache 2.0 weights. It scored strongly on maths benchmarks but the team documented clear failure modes.

  3. 2025
    • Jan 28, 2025 Qwen2.5-Max

      Qwen2.5-Max is Alibaba's large mixture-of-experts model, announced in January 2025. It was offered through Qwen Chat and the Alibaba Cloud API, and its weights were not released.

    • Mar 6, 2025 QwQ-32B

      A 32-billion-parameter open-weight reasoning model released by the Qwen team on 6 March 2025, trained mainly through scaled reinforcement learning and reported to perform comparably to the much larger DeepSeek-R1.

    • Apr 29, 2025 Qwen3

      Qwen3 is a family of large language models from Alibaba, released on 29 April 2025 under Apache 2.0. Its main feature is hybrid thinking, which lets one model reason step by step or answer directly.

    • Jul 21, 2025 Qwen3-2507

      Qwen3-2507 is a staged 2025 refresh of the open-weight Qwen3 models that split thinking and non-thinking into separate checkpoints, raised native context to 262,144 tokens, and scored far above the original non-thinking mode on published benchmarks.

    • Sep 2025 Qwen3-Max

      Qwen3-Max is Alibaba's proprietary trillion-parameter language model, released in September 2025 and served through Alibaba's API. It is the Max-tier flagship that followed Qwen2.5-Max.

    • Sep 2025 Qwen3-Next

      Qwen3-Next is an open-weight model from Alibaba's Qwen team, released in September 2025 under Apache 2.0. It has 80B total parameters, only 3B active per token, and a hybrid attention design later used in Qwen3.5.

  4. 2026
    • Feb 16, 2026 Qwen3.5 · This page

      Qwen3.5 is a natively multimodal model family from Alibaba released in February 2026. Its open flagship, Qwen3.5-397B-A17B, has 397B total parameters and 17B active per token, under Apache 2.0.

    • Apr 2, 2026 Qwen3.6

      A generation of Alibaba's Qwen models released in April 2026, spanning proprietary Plus and Max versions aimed at agentic coding and open-weight models such as Qwen3.6-35B-A3B, a 35B-parameter mixture-of-experts with 3B active parameters.

    • Aug 3, 2026 Qwen3.8

      Qwen3.8 is Alibaba's 2026 flagship line. Its open 2.4-trillion-parameter model, Qwen3.8-2.4T-A95B, released in August 2026, has 95B active parameters and a custom license.

20261

  1. ReleaseAlibaba releases natively multimodal Qwen3.5

    In February 2026 Alibaba released Qwen3.5, a natively multimodal model family whose open flagship has 397B parameters and was posted to Hugging Face on 16 February under Apache 2.0.

Developed by1

Alibaba Cloud

Builds on1

Qwen3-Next

Successor of1

Qwen3

Succeeded by1

Qwen3.6

Powers1

Qwen app (Tongyi)

Uses1

Mixture of experts (MoE)

Notes and sources

  1. Qwen/Qwen3.5-397B-A17Bhttps://huggingface.co/Qwen/Qwen3.5-397B-A17B
  2. Qwen3.5 modelshttps://huggingface.co/api/models?author=Qwen&search=Qwen3.5
  3. Qwenhttps://en.wikipedia.org/wiki/Qwen

Cite this page: Qwen3.5. TellWorld, https://www.tellworld.net/models/qwen3-5. Updated 2026-11-08