Gemma 4: Byte for byte, the most capable open models

Gemma 4：逐字节对比，最强大的开源模型

Today, we are introducing Gemma 4 — our most intelligent open models to date. Purpose-built for advanced reasoning and agentic workflows, Gemma 4 delivers an unprecedented level of intelligence-per-parameter. This breakthrough builds on incredible community momentum: since the launch of our first generation, developers have downloaded Gemma over 400 million times, building a vibrant Gemmaverse of more than 100,000 variants. We listened closely to what innovators need next to push the boundaries of AI, and Gemma 4 is our answer: breakthrough capabilities made widely accessible under an Apache 2.0 license.

今天，我们正式推出 Gemma 4——这是我们迄今为止最智能的开源模型。Gemma 4 专为高级推理和智能体（Agentic）工作流而构建，提供了前所未有的“单位参数智能水平”。这一突破建立在令人难以置信的社区势头之上：自第一代产品发布以来，开发者下载 Gemma 的次数已超过 4 亿次，构建了一个拥有超过 10 万个变体的活跃“Gemma 宇宙”。我们密切倾听了创新者在推动人工智能边界方面的需求，而 Gemma 4 就是我们的答案：在 Apache 2.0 许可下，让突破性的能力变得广泛可及。

Built from the same world-class research and technology as Gemini 3, Gemma 4 is the most capable model family you can run on your hardware. They complement our Gemini models, giving developers the industry’s most powerful combination of both open and proprietary tools.

Gemma 4 基于与 Gemini 3 相同的世界级研究和技术构建，是您可以在自己的硬件上运行的最强模型系列。它们与我们的 Gemini 模型相辅相成，为开发者提供了业界最强大的开源与闭源工具组合。

Industry-leading capabilities and mobile-first AI

行业领先的能力与移动优先的 AI

We are releasing Gemma 4 in four versatile sizes: Effective 2B (E2B), Effective 4B (E4B), 26B Mixture of Experts (MoE) and 31B Dense. The entire family moves beyond simple chat to handle complex logic and agentic workflows. Our larger models deliver state-of-the-art performance for their sizes, with the 31B model currently ranking as the #3 open model in the world on the industry-standard Arena AI text leaderboard, and the 26B model securing the #6 spot. There, Gemma 4 outcompetes models 20x its size. For developers, this new level of intelligence-per-parameter means achieving frontier-level capabilities with significantly less hardware overhead. At the edge, our E2B and E4B models redefine on-device utility, prioritizing multimodal capabilities, low-latency processing and seamless ecosystem integration over raw parameter count.

我们发布了四种不同规格的 Gemma 4 模型：有效参数 2B (E2B)、有效参数 4B (E4B)、26B 混合专家模型 (MoE) 以及 31B 稠密模型。整个系列超越了简单的聊天功能，能够处理复杂的逻辑和智能体工作流。我们的大型模型在同等规模下提供了顶尖的性能，其中 31B 模型目前在行业标准的 Arena AI 文本排行榜上排名全球第三，26B 模型位列第六。在榜单中，Gemma 4 的表现超越了规模为其 20 倍的模型。对于开发者而言，这种全新的“单位参数智能水平”意味着可以用更少的硬件开销实现前沿级的能力。在边缘侧，我们的 E2B 和 E4B 模型重新定义了端侧实用性，相比单纯的参数规模，它们更优先考虑多模态能力、低延迟处理以及无缝的生态系统集成。

Powerful, accessible, open

强大、易用且开放

To power the next generation of pioneering research and products, we’ve sized the Gemma 4 models specifically to run and fine-tune efficiently on hardware — from billions of Android devices worldwide, to laptop GPUs, all the way up to developer workstations and accelerators. By using these highly optimized models, you can fine-tune Gemma 4 to achieve state-of-the-art performance on your specific tasks. We’ve already seen incredible success with this approach; for instance, INSAIT created a pioneering Bulgarian-first language model (BgGPT), and we worked with Yale University on Cell2Sentence-Scale to discover new pathways for cancer therapy, among many others.

为了赋能下一代开创性的研究和产品，我们专门调整了 Gemma 4 模型的大小，使其能够在各种硬件上高效运行和微调——从全球数十亿台 Android 设备到笔记本电脑 GPU，再到开发者工作站和加速器。通过使用这些高度优化的模型，您可以对 Gemma 4 进行微调，从而在特定任务上实现顶尖性能。我们已经见证了这种方法取得的巨大成功；例如，INSAIT 创建了保加利亚首个开创性语言模型 (BgGPT)，我们还与耶鲁大学在 Cell2Sentence-Scale 项目上合作，探索癌症治疗的新途径，等等。

Here is what makes Gemma 4 our most capable open model family yet:

以下是 Gemma 4 成为我们迄今最强开源模型系列的原因：

Advanced reasoning: Capable of multi-step planning and deep logic, Gemma 4 demonstrates significant improvements in math and instruction-following benchmarks that require it.
高级推理： 具备多步规划和深度逻辑能力，Gemma 4 在需要此类能力的数学和指令遵循基准测试中表现出显著提升。
Agentic workflows: Native support for function-calling, structured JSON output, and native system instructions enables you to build autonomous agents that can interact with different tools and APIs and execute workflows reliably.
智能体工作流： 原生支持函数调用、结构化 JSON 输出和原生系统指令，使您能够构建能够与不同工具和 API 交互并可靠执行工作流的自主智能体。
Code generation: Gemma 4 supports high-quality offline code, turning your workstation into a local-first AI code assistant.
代码生成： Gemma 4 支持高质量的离线代码生成，将您的工作站变成一个本地优先的 AI 代码助手。
Vision and audio: All models natively process video and images, supporting variable resolutions, and excelling at visual tasks like OCR and chart understanding. Additionally, the E2B and E4B models feature native audio input for speech recognition and understanding.
视觉与音频： 所有模型均可原生处理视频和图像，支持多种分辨率，并在 OCR 和图表理解等视觉任务中表现出色。此外，E2B 和 E4B 模型还具备用于语音识别和理解的原生音频输入功能。
Longer context: Process long-form content seamlessly. The edge models feature a 128K context window, while the larger models offer up to 256K, allowing you to pass repositories or long documents in a single prompt.
更长上下文： 无缝处理长篇内容。边缘模型具有 128K 上下文窗口，而大型模型则提供高达 256K 的窗口，允许您在单个提示词中传入代码库或长文档。
140+ languages: Natively trained on over 140 languages, Gemma 4 helps developers build inclusive, high-performance applications for a global audience.
140 多种语言： 原生训练涵盖 140 多种语言，Gemma 4 帮助开发者为全球用户构建包容性强、高性能的应用程序。