Gemma 4 E2B (Reasoning)
GoogleGemma开源权重Apache 2.0 · 商用许可
描述
Gemma 4 E2B is Google DeepMind's smallest multimodal model with 2.3 billion effective parameters (5.1B with embeddings) and a 128K context window. Supports image, text, and audio inputs. Designed for on-device and edge deployment with Per-Layer Embeddings for efficient inference.
发布日期
2026-04-02
参数规模
5.1B
上下文长度
—
支持模态
—
能力雷达图
40
general
0
coding
50
reasoning
34
science估算
30
agents
40
multimodal
缺少专门科学评测时,Science 由 LLM Stats 科学得分或推理能力估算。
排行榜排名
基准测试分数 (LLM Stats)
(LLM Stats (zeroeval))Agents
t2-bench
29.4%自报
Healthcare
MedXpertQA
23.5%自报
Language
MMMLU
67.4%自报
MMLU-Pro
60.0%自报
Long Context
MRCR v2 (8-needle)
19.1%自报
Math
MathVision
52.4%自报
AIME 2026
37.5%自报
Reasoning
LiveCodeBench v6
44.0%自报
GPQANYU + Cohere + Anthropic (2023)
43.4%自报
BIG-Bench Extra Hard
21.9%自报
Vision
MMMU-Pro
44.2%自报
AA 评测指数
(Artificial Analysis)暂无 AA 评测数据
LLM Stats 分类评分
(LLM Stats (zeroeval))Legal60
Finance60
Language50
Math50
Multimodal40
Physics40
Reasoning40
General40
Healthcare40
Biology40
Chemistry40
Vision40
Agents30
Tool Calling30
Long Context20
定价
暂无定价数据
速度
暂无速度数据
供应商价格排行
暂无提供商数据