See key specs and per-benchmark scores for each model/mode. Scroll horizontally for all columns. 当前对比 2 个模型的评测数据与核心参数。
Gemma 4 31B（稠密旗舰模型）
Gemma 3 - 27B (IT)
Compare benchmark results across thinking modes and tool usage.
Data sourced primarily from official releases (GitHub, Hugging Face, papers), then benchmark leaderboards, then third-party evaluators. Learn about our data methodology
Best Overall
—
Best Single
—
Thinking Mode (Default)
Gemma 4 31B · 0 Modality support
Higher is usually better; “—” means no score.
Complete scores for each model/mode across selected benchmarks.
Feature compare
Licensing, MoE architecture, and multi-modality support.
| Features & specs | GE Gemma 4 31BDeepMind | GE Gemma 3 - 27B (IT)Google Deep Mind |
|---|---|---|
Model snapshots | ||
Organization | DeepMind | Google Deep Mind |
模型全名 | Gemma 4 31B（稠密旗舰模型） | Gemma 3 - 27B (IT) |
模型简介 | Not provided | Not provided |
模型类型 | 聊天大模型 | 聊天大模型 |
模型代号 | google-gemma-4-31b | gemma-3-27b-it |
Release | 2026-04-02 | 2025-03-12 |
MoE | No | No |
规格与性能 | ||
Context length | 256K | 128K |
Parameters | 31 | 270 |
激活参数量 | 31 | Not provided |
模型规模 | 3b | 34b |
模型大小 | Not provided | 54.8GB |
推理速度 | Not provided | Not provided |
推理等级 | Not provided | Not provided |
最大输出 | 32768 | Not provided |
Supported modes | No mode data | No mode data |
开源与许可 | ||
Code Open Source | Not provided | Not provided |
Weights Open Source | Not provided | Not provided |
Commercial use | 免费商用授权 | 免费商用授权 |
Modality support | ||
Text Input/Output | Not provided | Not provided |
Image Input/Output | Not provided | Not provided |
Audio Input/Output | Not provided | Not provided |
Video Input/Output | Not provided | Not provided |
Embedding Input/Output | Not provided | Not provided |
API 接口详情 | ||
Text 价格 | Not provided | Not provided |
Image API pricing | Not provided | Not provided |
Audio API pricing | Not provided | Not provided |
Video API pricing | Not provided | Not provided |
Embedding API pricing | Not provided | Not provided |
Resources | ||
GitHub | Not provided | Not provided |
Hugging Face | Model Page | Model Page |
Official Page | Not provided | Not provided |
Guides | Not provided | Not provided |
Papers | Gemma 4: Byte for byte, the most capable open models | Gemma 3 Technical Report |
DataLearnerAI | 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 | Google开源第三代Gemma-3系列模型：支持多模态、最多128K输入，其中Gemma 3-27B在大模型匿名竞技场得分超过了Qwen2.5-Max |
API pricing
Side-by-side input/output token pricing
