EmbeddingGemma 2

Google·US·Released 2026-10-06

ClosedEmbeddingEmbedding

An embedding model from Google in the Gemma family, released October 2026 with closed weights.

—context

—per 1M tokens (in / out)

—parameters

55stories in our feed

Specs

hover a value to see its source

Context window—

Max output—

Parameters—

Knowledge cutoff—

Licence—

Commercial useUnknown

Input—

Output—

Familygemma

API price

USD per 1M tokens

No public API price yet.

News timeline

from the AITOP feed, one story per eventAll 55 →

2026-10-21

Gemini 4 Argon 一句话生成本地开源健康 AI 应用Paul Couvert

2026-10-17

Cohere Embed 5 Fast Beats Google EmbeddingGemma 2 on 163 DatasetsCohere

2026-10-16

Google 与新加坡 GovTech 合作，用微调 Gemma 模型提醒用户诈骗短信Blognone

2026-10-15

Gemma 推出全新官方文档站 gemma.aiPhilipp Schmid

EmbeddingGemma 2 fine-tune achieves SOTA Vietnamese ASR for US$78AIGCLINK

开源工具 DigUp 基于 EmbeddingGemma 2 实现 Mac 本地跨模态搜索小互

2026-10-14

Inworld Inference API targets consumer AI appsTestingCatalog

EmbeddingGemma 2 powers local macOS semantic search toolAIGCLINK

2026-10-13

LiquidAI LFM2.5-24B-A2B 本地对决 Gemma 4 26B 全胜@atomic_chat_hq

Anthropic 红队研究：AI 可借 15 台廉价电视盒子实现自我复制小互

2026-10-10

Google 推出 Mac 免费 AI 笔记应用，本地运行 Embedding Gemma 2Shubham Saboo

2026-10-08

ATOMIC开源桌面应用支持多平台运行TestingCatalog

2026-10-07

谷歌推出 macOS 应用 AI Edge Foresight，基于 EmbeddingGemma 2 离线生成会议纪要IT之家

Google EmbeddingGemma 2 Model ReleasedAIGCLINK

2026-10-05

MIT Paper Finds LLM Agents Bid Better With Simpler Prompt Formatsrohanpaul_ai

2026-10-02

Gemma 4 开发者代理竞赛启动Google Research

语言模型自修复机制研究：增益与反作用力arXiv cs.LG

2026-09-29

LLM Post-training Robust to Erroneous RewardsTanishq Abraham (论文推介)

2026-09-28

论文：自监督置信度微调让推理模型生成 token 减少 25%arXiv cs.LG

2026-09-25

面向加纳三语青少年健康语音识别的基准测试与领域适配研究arXiv cs.AI

2026-09-24

Google Gemma 4 Developer Agent 竞赛开启，总奖金 10 万美元Kaggle

2026-09-16

SiFive and AMD partner to optimize ROCm on RISC-V processorsIT之家

2026-09-11

How LLMs Retrieve and Use Internal KnowledgearXiv cs.AI

2026-09-04

ESPO：错误结构化提示优化方法arXiv cs.AI

2026-08-20

Gemma 模型下载量突破 10 亿ollama

Gemma resources list on GitHubPhilipp Schmid

2026-08-14

Hugging Face报告：Qwen领跑本地推理，小模型主导实际使用阿里通义 Qwen

2026-08-06

多智能体协作是好事，Clement Delangue晒出149个Gemma实验Clement Delangue

2026-07-29

Evaluating VLMs for Agent-Driven Geometry Clipping Detection in Game QAarXiv cs.AI

2026-07-25

Google AI Edge：小模型在边缘设备上的实际部署AI Engineer

Sundar Pichai代表Google支持开源，提及Gemma模型Sundar Pichai

Demis Hassabis和NVIDIA联合强调开放模型对AI生态的关键作用Demis Hassabis

在 8 美元 ESP32-S3 上运行近 2900 万参数 LLM，利用 Gemma Per-Layer Embeddings 将参数存于 FlashGeek

2026-07-21

压力下的逻辑判断：用学习软前缀诊断三段论稳定性arXiv cs.AI