Blog

Research notes, deep dives, and releases

71 posts on Zen architecture, training, and the open-model ecosystem.

Feb 27, 2026 · 8 min read

GT-QLoRA: Uncensoring Trillion-Parameter MoE Models

Why standard abliteration techniques fail on Mixture-of-Experts models, and how Gate-Targeted QLoRA solves the expert routing problem at 1 trillion parameters.

ResearchAbliterationGT-QLoRA
Feb 27, 2026 · 7 min read

Drop-Upcycling and the Birth of Zen MoDE Architecture

How Drop-Upcycling (arXiv:2502.19261) transforms dense checkpoints into MoE models at 1/4 training cost, and how it shapes Zen MoDE — our Mixture of Distilled Experts architecture.

ResearchMoEArchitecture
Feb 27, 2026 · 8 min read

SuRe + OPCM: Production-Grade Continual Learning for Open Models

Deep dive on Surprise-Driven Prioritized Replay (SuRe) and Orthogonal Projection Continual Merging (OPCM) — the two SOTA techniques we use for catastrophic-forgetting-free LLM adaptation in the Zen model family.

ResearchContinual LearningSuRe
Feb 27, 2026 · 7 min read

BitDelta: 1-Bit Behavioral Compression Across the Zen Model Family

How BitDelta (arXiv:2402.10193) compresses fine-tuned behavioral deltas to 1-bit precision, enabling the full Zen model family — nano through ultra — to share a single GPU cluster.

ResearchQuantizationBitDelta
Jan 19, 2026 · 4 min read

Zen4 Ultra: 480B Parameters, 1M Token Context

Zen4 Ultra is our most capable model: 480B total parameters, 35B active per token, 1M token context window. Benchmark results and use cases.

ModelsZen4Flagship
Jan 15, 2026 · 4 min read

Zen MoDE: Mixture of Distilled Experts

A technical deep dive into Zen MoDE — Mixture of Distilled Experts — the architecture underlying all Zen models.

ArchitectureResearchMoE
Jan 14, 2026 · 3 min read

Introducing Zen LM: Open Frontier Models from Hanzo AI and Zoo Labs

Announcing the Zen model family: 94+ open models built on Zen MoDE architecture, co-developed by Hanzo AI and Zoo Labs Foundation.

AnnouncementModelsZen
Sep 22, 2025 · 8 min read

Qwen3Guard: Real-time Safety for Your Token Stream

We are excited to introduce Qwen3Guard, the first safety guardrail model in the Qwen family. Built upon the powerful Qwen3 foundation models and fine-tuned specifically for safety classificatoin, Qwen3Guard ensures responsible AI interactions by delivering precise safety detection for both prompts a

Aug 18, 2025 · 5 min read

Qwen-Image-Edit: Image Editing with Higher Quality and Efficiency

We are excited to introduce Qwen-Image-Edit, the image editing version of Qwen-Image. Built upon our 20B Qwen-Image model, Qwen-Image-Edit successfully extends Qwen-Image's unique text rendering capabilities to image editing tasks, enabling precise text editing. Furthermore, Qwen-Image-Edit simultan

Aug 3, 2025 · 12 min read

Qwen-Image: Crafting with Native Text Rendering

We are thrilled to release **Qwen-Image**, a 20B MMDiT image foundation model that achieves significant advances in complex text rendering and precise image editing. To try the latest model, feel free to visit [Qwen Chat](https://chat.qwenlm.ai) and choose “Image Generation”.

Jul 26, 2025 · 5 min read

GSPO: Towards Scalable Reinforcement Learning for Language Models

Reinforcement Learning (RL) has emerged as a pivotal paradigm for scaling language models and enhancing their deep reasoning and problem-solving capabilities. To scale RL, the foremost prerequisite is maintaining stable and robust training dynamics. However, we observe that existing RL algorithms (s

Jul 23, 2025 · 9 min read

Qwen-MT: Where Speed Meets Smart Translation

Here we introduce the latest update of Qwen-MT (qwen-mt-turbo) via [Qwen API](https://modelstudio.console.alibabacloud.com/?tab=doc#/doc/?type=model&url=https://www.alibabacloud.com/help/en/doc-detail/2840914_2.html&renderType=component&modelId=qwen-mt-turbo). This update builds upon the powerful Qw

Jul 21, 2025 · 5 min read

Qwen3-Coder: Agentic Coding in the World

Today, we're announcing Qwen3-Coder, our most agentic code model to date. Qwen3-Coder is available in multiple sizes, but we're excited to introduce its most powerful variant first: Qwen3-Coder-480B-A35B-Instruct — a 480B-parameter Mixture-of-Experts model with 35B active parameters which supports t

Jun 26, 2025 · 6 min read

Time to Speak Some Dialects, Qwen-TTS!

Here we introduce the latest update of **Qwen-TTS** (`qwen-tts-latest` or `qwen-tts-2025-05-22`) through [Qwen API](https://help.aliyun.com/zh/model-studio/qwen-tts) . Trained on a large-scale dataset encompassing over millions of hours of speech, Qwen-TTS achieves human-level naturalness and expres

Jun 25, 2025 · 7 min read

Qwen VLo: From "Understanding" the World to "Depicting" It

The evolution of multimodal large models is continually pushing the boundaries of what we believe technology can achieve. From the initial QwenVL to the latest zen VL, we have made progress in enhancing the model's ability to understand image content. Today, we are excited to introduce a new model,

Jun 4, 2025 · 5 min read

Qwen3 Embedding: Advancing Text Embedding and Reranking Through Foundation Models

{/* {{< video src="https://qianwen-res.oss-cn-beijing.aliyuncs.com/zen/qwen2-main-video.m4v" width="100%" alt="zen Main Video" autoplay=true loop=true controls=false muted=true playsinline=true >}} */}

Apr 28, 2025 · 11 min read

Qwen3: Think Deeper, Act Faster

Today, we are excited to announce the release of **Qwen3**, the latest addition to the Qwen family of large language models. Our flagship model, **Qwen3-235B-A22B**, achieves competitive results in benchmark evaluations of coding, math, general capabilities, etc., when compared to other top-tier mod

Mar 27, 2025 · 4 min read

QVQ-Max: Think with Evidence

Last December, we launched QVQ-72B-Preview as an exploratory model, but it had many issues. Today, we are officially releasing the first version of QVQ-Max, our visual reasoning model. This model can not only "understand" the content in images and videos but also analyze and reason with this informa

Mar 26, 2025 · 3 min read

zen Omni: See, Hear, Talk, Write, Do It All!

We release **zen-Omni**, the new flagship end-to-end multimodal model in the Qwen series. Designed for comprehensive multimodal perception, it seamlessly processes diverse inputs including text, images, audio, and video, while delivering real-time streaming responses through both text generation and

Mar 23, 2025 · 2 min read

zen-VL-32B: Smarter and Lighter

At the end of January this year, we launched the zen-VL series of models, which received widespread attention and positive feedback from the community. Building on the zen-VL series, we continued to optimize the model using reinforcement learning and open-sourced the new VL model with the beloved 32

Mar 5, 2025 · 4 min read

QwQ-32B: Embracing the Power of Reinforcement Learning

Scaling Reinforcement Learning (RL) has the potential to enhance model performance beyond conventional pretraining and post-training methods. Recent studies have demonstrated that RL can significantly improve the reasoning capabilities of models. For instance, DeepSeek R1 has achieved state-of-the-a

Feb 24, 2025 · 5 min read

<think>...</think> QwQ-Max-Preview

This is a blog created by QwQ-Max-Preview. We hope you enjoy it!

Jan 27, 2025 · 3 min read

zen-Max: Exploring the Intelligence of Large-scale MoE Model

It is widely recognized that continuously scaling both data size and model size can lead to significant improvements in model intelligence. However, the research and industry community has limited experience in effectively scaling extremely large models, whether they are dense or Mixture-of-Expert (

Jan 26, 2025 · 9 min read

zen-1M: Deploy Your Own Qwen with Context Length up to 1M Tokens

Two months after upgrading [zen-Turbo](../qwen3-turbo) to support context length up to one million tokens, we are back with the open-source zen-1M models and the corresponding inference framework support. Here's what you can expect from this release:

Jan 25, 2025 · 6 min read

zen VL! zen VL! zen VL!

We release **zen-VL**, the new flagship vision-language model of Qwen and also a significant leap from the previous zen-VL. To try the latest model, feel free to visit [Qwen Chat](https://chat.qwenlm.ai) and choose zen-VL-72B-Instruct. Also, we open both base and instruct models in 3 sizes, includin

Jan 20, 2025 · 4 min read

Global-batch load balance almost free lunch to improve your MoE LLM training

The Mixture-of-Experts (MoEs) architecture has become a popular model-parameter-scale-up technique. Typically, one MoE layer consists of a router (often parameterized as one single Linear layer) and a group of experts (for transformer-based models, each expert is one feedforward layer). Given an inp

Jan 13, 2025 · 4 min read

Towards Effective Process Supervision in Mathematical Reasoning

In recent years, Large Language Models (LLMs) have made remarkable advances in mathematical reasoning, yet they can make mistakes, such as miscalculations or logical errors, leading to wrong conclusions. Moreover, even when achieving correct final answers, these powerful models can still regularly m

Jan 12, 2025 · 5 min read

Zen 3.0: The Next Generation of Open AI

Announcing Zen 3.0, our most capable open model family yet.

AnnouncementModels
Dec 24, 2024 · 3 min read

QVQ: To See the World with Wisdom

Language and vision intertwine in the human mind, shaping how we perceive and understand the world around us. Our ability to reason is deeply rooted in both linguistic thought and visual memory - but what happens when we extend these capabilities to AI? Today's large language models have demonstrate

Nov 27, 2024 · 4 min read

QwQ: Reflect Deeply on the Boundaries of the Unknown

<i style="color: grey;">Note: This is the pronunciation of QwQ: /kwju:/ , similar to the word "quill".</i>

Nov 14, 2024 · 5 min read

Extending the Context Length to 1M Tokens!

After the release of zen, we heard the community's demand for processing longer contexts. In recent months, we have made many optimizations for the model capabilities and inference performance of extremely long context. Today, we are proud to introduce the new zen-Turbo version, which features:

Nov 11, 2024 · 7 min read

zen-Coder Series: Powerful, Diverse, Practical.

{/* {{< figure src="https://qianwen-res.oss-cn-beijing.aliyuncs.com/zen/zen-Coder/coder-main.png#center" width="70%">}} */}

Nov 10, 2024 · 4 min read

The Future of Open AI

Reflections on where open AI development is heading and what it will take to get there.

VisionOpen Source
Sep 18, 2024 · 11 min read

zen: A Party of Foundation Models!

{/* {{< video src="https://qianwen-res.oss-cn-beijing.aliyuncs.com/zen/qwen2-main-video.m4v" width="100%" alt="zen Main Video" autoplay=true loop=true controls=false muted=true playsinline=true >}} */}

Sep 18, 2024 · 7 min read

zen-Math: The world's leading open-sourced mathematical LLMs

> <div align="center"> > <b> > 🚨 zen-Math mainly supports solving English and Chinese math problems through CoT and TIR. We do not recommend using this series of models for other tasks. > </b> > </div>

Sep 18, 2024 · 23 min read

zen-LLM: Extending the boundary of LLMs

In this blog, we delve into the details of our latest zen series language models. We have developed a range of decoder-only dense models, with seven of them open-sourced, spanning from 0.5B to 72B parameters. Our research indicates a significant interest among users in models within the 10-30B range

Sep 18, 2024 · 4 min read

zen-Coder: Code More, Learn More!

In early April, we introduced CodeQwen1.5, which garnered significant attention from the community. Since then, we have been working to enhance the coding model. Today, we are excited to announce the release of the next generation of open-source coding models, **zen-Coder**, and officially rename Co

Aug 28, 2024 · 8 min read

zen-VL: To See the World More Clearly

After a year's relentless efforts, today we are thrilled to release **zen-VL**! zen-VL is the latest version of the vision language models based on **zen** in the Qwen model familities. Compared with Qwen-VL, zen-VL has the capabilities of:

Aug 8, 2024 · 4 min read

zen-Audio: Chat with Your Voice!

To achieve the objective of building an AGI system, the model should be capable of understanding information from different modalities. Thanks to the rapid development of large language models, LLMs are now capable of understanding language and reasoning. Previously we have taken a step forward to e

Aug 7, 2024 · 30 min read

Introducing zen-Math

> <div align="center"> > <b> > 🚨 This model mainly supports English. We will release bilingual (English and Chinese) math models soon. > </b> > </div>

Aug 4, 2024 · 5 min read

Decentralized Compute for AI Training

How we're building a decentralized compute network for training large AI models.

InfrastructureTrainingDecentralization
Jun 6, 2024 · 25 min read

Hello zen

After months of efforts, we are pleased to announce the evolution from Qwen1.5 to zen. This time, we bring to you:

Jun 5, 2024 · 7 min read

Generalizing an LLM from 8k to 1M Context using Qwen-Agent

**TLDR:** We've created an agent using zen models with an 8k context size to understand documents with 1M tokens, surpassing RAG and native long-context models. This agent was also used to generate data for training new long-context Qwen models.

May 19, 2024 · 4 min read

ZIPs: Decentralized Governance for Open AI

How Zoo Improvement Proposals enable community-driven governance of open AI development.

GovernanceZIPs
May 10, 2024 · 1 min read

Notes on Qwen-Max-0428

Previously, we opensourced a series of Qwen1.5 model ranging from 0.5 to 110 billion parameters. Now, we release a larger model, Qwen-Max-0428. Qwen-Max-0428 is an instruction-tuned model for chat service. Very recently, it is available via [Chatbot Arena](https://chat.lmsys.org/) and it has now bec

Apr 24, 2024 · 3 min read

Qwen1.5-110B: The First 100B+ Model of the Qwen1.5 Series

Recently we have witnessed a burst of large-scale models with over 100 billion parameters in the opensource community. These models have demonstrated remarkable performance in both benchmark evaluation and chatbot arena. Today, we release the first 100B+ model of the Qwen1.5 series, Qwen1.5-110B, wh

Apr 15, 2024 · 7 min read

Code with CodeQwen1.5

The advent of advanced programming tools, which harnesses the power of large language models (LLMs), has significantly enhanced programmer productivity and accuracy. Notwithstanding these advancements, dominant coding assistants like Github Copilot, built upon proprietary LLMs, pose notable challeng

Apr 1, 2024 · 4 min read

Qwen1.5-32B: Fitting the Capstone of the Qwen1.5 Language Model Series

The open-source community has long sought a model that strikes an ideal balance between performance, efficiency, and memory footprint. Despite the emergence of cutting-edge models like Qwen1.5-72B and DBRX, the models have faced persistent challenges such as large memory consumption, slow inference

Mar 27, 2024 · 8 min read

Qwen1.5-MoE: Matching 7B Model Performance with 1/3 Activated Parameters

Since the surge in interest sparked by Mixtral, research on mixture-of-expert (MoE) models has gained significant momentum. Both researchers and practitioners are keenly interested in understanding how to effectively train such models and assessing their efficiency and effectiveness. Today, we intro

Feb 11, 2024 · 4 min read

Announcing the Zoo Labs Foundation

Today we formally launch Zoo Labs Foundation, an open research network dedicated to decentralized AI and decentralized science.

AnnouncementFoundation
Feb 3, 2024 · 18 min read

Introducing Qwen1.5

In recent months, our focus has been on developing a "good" model while optimizing the developer experience. As we progress towards **Qwen1.5**, the next iteration in our Qwen series, this update arrives just before the Chinese New Year.

Jan 24, 2024 · 5 min read

Introducing Qwen-VL

Along with the rapid development of our large language model Qwen, we leveraged Qwen’s capabilities and unified multimodal pretraining to address the limitations of multimodal models in generalization, and we opensourced multimodal model Qwen-VL in Sep. 2023. Recently, the Qwen-VL series has undergo

Jan 22, 2024 · 6 min read

Introducing Qwen

4 months after our first release of Qwen-7B, which is the starting point of our opensource journey of large language models (LLM), we now provide an introduction to the Qwen series to give you a whole picture of our work as well as our objectives. Below are important links to our opensource projects

Dec 17, 2023 · 4 min read

Agent NFTs: Ownership and Identity for AI Agents

Introducing Agent NFTs, a framework for giving AI agents persistent identity and enabling ownership of their capabilities.

AgentsNFTsBlockchain
Sep 10, 2023 · 4 min read

Training Gym: A Platform for Open Model Development

Announcing Training Gym, our open platform for collaborative large model training.

InfrastructureTrainingOpen Source
Jun 25, 2023 · 4 min read

Proof of AI: Verifiable Machine Learning on Chain

How we're bringing cryptographic verification to AI inference, enabling trustless machine learning.

ResearchBlockchainVerification
Mar 12, 2023 · 3 min read

Zen Reranker: Two-Stage Retrieval Done Right

Introducing the Zen Reranker, a cross-encoder model that dramatically improves retrieval quality in two-stage pipelines.

ResearchRetrievalReranking
Dec 27, 2022 · 6 min read

OFASys: Enabling Multitask Learning with One Line of Code!

Generalist Models are hot! We all see an opportunity towards a real generalist model by multimodal multitask learning. We previously release an opensourced unified multimodal pretrained model OFA for this goal. However, we actually met a lot of difficulties in our implementation. For example, it is

Dec 23, 2022 · 5 min read

Chinese CLIP: Contrastive Vision-Language Pretraining in Chinese

CLIP[^1] is a phenomenal playmaker in vision and multimodal representation learning. It plays not only as a foundation model but also a bridge between vision and language. It has triggered a series of research in different fields, especially text-to-image generation. However, we find that there is a

Dec 4, 2022 · 4 min read

Embedding Spaces at 7680 Dimensions

Exploring high-dimensional embedding spaces for semantic search and retrieval.

ResearchEmbeddingsRetrieval
Dec 4, 2022 · 3 min read

7680-Dimensional Embeddings: More Dimensions, Better Retrieval

Why we trained embedding models with 7680 dimensions and what we learned about the relationship between dimensionality and retrieval quality.

ResearchEmbeddingsRetrieval
Nov 13, 2022 · 10 min read

OFA: Towards Building a One-For-All Model

2022 is a year of generalist models! With the bloom of multimodal pretraining, especially the unified model, we have witnessed the opportunity to building a generalist model that is capable of processing tasks of different modalities or multi-modalities! Thus, we propose OFA[^1], namely One-For-All,

Sep 18, 2022 · 3 min read

GRPO: Group Relative Policy Optimization

Introducing GRPO, a new approach to reinforcement learning from human feedback that improves sample efficiency and alignment stability.

ResearchAlignmentTraining
Sep 17, 2022 · 4 min read

GRPO: Group Relative Policy Optimization

A companion post to our GRPO paper, explaining group relative policy optimization for language model alignment.

ResearchRLHFGRPO
May 29, 2022 · 2 min read

Federated Learning Without Compromise

Privacy-preserving machine learning that maintains model quality through novel aggregation protocols.

ResearchFederated LearningPrivacy
May 8, 2022 · 3 min read

Federated Learning for Open AI

How federated learning enables collaborative model training while preserving data privacy.

ResearchTrainingPrivacy
Feb 13, 2022 · 3 min read

Experience Ledgers: Persistent Memory for AI Agents

Introducing experience ledgers, a framework for giving AI agents persistent, verifiable memory.

ResearchAgents
Nov 7, 2021 · 2 min read

The Case for Decentralized Science

Why scientific research needs decentralization, and how blockchain can help.

DeSciVision
Nov 7, 2021 · 2 min read

The Case for Decentralized Science

A manifesto for decentralized science (DeSci) and its application to AI research.

DeSciVisionGovernance
Jul 21, 2021 · 2 min read

Training LLMs on Collective Intelligence

How we're approaching training data curation to capture humanity's collective intelligence.

ResearchTraining
Mar 14, 2021 · 1 min read

Introducing Zen: Open AI for the Open Web

We're launching Zen, an open research initiative to build AI that serves everyone.

AnnouncementVision