DataLearner logoDataLearnerAI
Latest AI Insights
Model Leaderboards
Benchmarks
Model Directory
Model Comparison
Resource Center
Tools
LanguageEnglish
DataLearner logoDataLearner AI

A knowledge platform focused on LLM benchmarking, datasets, and practical instruction with continuously updated capability maps.

Products

  • Leaderboards
  • Model comparison
  • Datasets

Resources

  • Tutorials
  • Editorial
  • Tool directory

Company

  • About
  • Privacy policy
  • Data methodology
  • Contact

© 2026 DataLearner AI. DataLearner curates industry data and case studies so researchers, enterprises, and developers can rely on trustworthy intelligence.

Privacy policyTerms of service
Page navigation
目录
Model catalogGemma 4 26B A4B
GE

Gemma 4 26B A4B

聊天大模型

Gemma 4 26B A4B(混合专家模型)

Release date: 2026-04更新于: 2026-04-05 10:27:43知识截止: 2025-12564
Live demoGitHubHugging FaceCompare
Parameters
25.2亿
Context length
256K
Chinese support
Not supported
Reasoning ability

Gemma 4 26B A4B(混合专家模型) is an AI model published by DeepMind, released on 2026-04, for 聊天大模型, with 25.2B parameters, and 256K tokens context length, under the Apache 2.0 license.

Data sourced primarily from official releases (GitHub, Hugging Face, papers), then benchmark leaderboards, then third-party evaluators. Learn about our data methodology

Gemma 4 26B A4B

Model basics

Reasoning traces
Supported
Thinking modes
Thinking modes not supported
Context length
256K tokens
Max output length
32768 tokens
Model type
聊天大模型
Release date
2026-04
Model file size
No data
MoE architecture
Yes
Total params / Active params
25.2B / 4B
Knowledge cutoff
2025-12
Gemma 4 26B A4B

Open source & experience

Code license
Apache 2.0
Weights license
Apache 2.0- 免费商用授权
GitHub repo
GitHub link unavailable
Hugging Face
https://huggingface.co/google/gemma-4-26b-a4b
Live demo
https://ai.google.dev/gemma
Gemma 4 26B A4B

Official resources

Paper
No paper available
DataLearnerAI blog
Google Gemma 4 正式开源:Apache 2.0 协议、手机端可运行、原生支持多模态和 Agent 工作流
Gemma 4 26B A4B

API details

API speed
No data
No public API pricing yet.
Gemma 4 26B A4B

Benchmark Results

Gemma 4 26B A4B currently shows benchmark results led by LiveCodeBench (25 / 109, score 77.10), MMLU Pro (37 / 116, score 82.60), GPQA Diamond (54 / 166, score 82.30). This page also consolidates core specs, context limits, and API pricing so you can evaluate the model from benchmark results and deployment constraints together.

Thinking
All modesThinking
Tool usage
All modesWith toolsNo tools
Internet
All modesOfflineInternet enabled

综合评估

4 evaluations
Benchmark / mode
Score
Rank/total
MMLU Pro
On
82.60
37 / 116
GPQA Diamond
On
82.30
54 / 166
HLE
On
8.70
109 / 128
HLE
OnToolsInternet
17.20
89 / 128

编程与软件工程

1 evaluations
Benchmark / mode
Score
Rank/total
LiveCodeBench
On
77.10
25 / 109

Agent能力评测

1 evaluations
Benchmark / mode
Score
Rank/total
τ²-Bench
OnTools
68.20
25 / 40

文本向量检索

1 evaluations
Benchmark / mode
Score
Rank/total
Context Arena
On
44.10
1 / 1

数学推理

1 evaluations
Benchmark / mode
Score
Rank/total
AIME 2026
On
88.30
11 / 11
View benchmark analysisCompare with other models
Gemma 4 26B A4B

Publisher

DeepMind
DeepMind
View publisher details
Gemma 4 26B A4B(混合专家模型)

Model Overview

Gemma 4 26B A4B 是 Google DeepMind 于 2026 年 4 月 2 日发布的新一代开源大语言模型系列中的高性能混合专家(MoE)型号。该模型采用 Apache 2.0 许可证,基于 Gemini 3 同源技术构建,原生支持多模态输入(文本、图像和视频)。模型采用 MoE 架构,包含 128 个专家,每次激活 8 个加 1 个共享专家,总参数 252 亿,推理时仅激活 38 亿参数。支持 256K 上下文窗口,支持 140 多种语言。在 Arena AI 开源排行榜中位列第六,Elo 评分 1441,推理速度接近 4B 模型,质量远超 4B 水平,性价比极高。相比前代 Gemma 3,数学、编程和推理能力实现代际跨越,同时新增了原生函数调用和结构化 JSON 输出,支持构建自主智能体工作流。

DataLearner 官方微信

欢迎关注 DataLearner 官方微信,获得最新 AI 技术推送

DataLearner 官方微信二维码