GEMMA4 4

Google

Gemma4_4b_128k_b1_2core_1chip()

GEMMA Gemma4 4 gguf w4a8 131,072 tokens M50 Testing 2.1.0

Download Model

https://llmoss.houmoai.com/application_software/models/general/1.4.0/HiModel_xh2_gemma4-e4b_320_128k_b1_1chip_2cores_v1.4.0_20260708.gguf

Download Now

Build: 1.4.0

Note: Models built on v1.3 require NPU driver 1.3 or above to work properly.

Note: Models for inference framework 2.1.0 require inference framework 2.1.0 or above to work properly.

MMProj (Vision Projector)

https://llmoss.houmoai.com/application_software/models/general/1.4.0/mmproj_xh2_gemma4-e4b_320_2cores_v1.4.0_20260722.gguf

Download

4

Parameters

131,072

Tokens

gguf

Format

—

Target

1

Batch

2

CPU Cores

Basic Info

Model ID / 模型 ID	Gemma4_4b_128k_b1_2core_1chip()
Vendor / 厂商	Google
Name (CN) / 中文名	GEMMA4 4
Name (EN) / 英文名	GEMMA4 4
Series / 系列	GEMMA
Type / 类型	Gemma4

Tech Specs

Format / 格式	gguf
Parameters / 参数	4
Quantization / 量化	w4a8
Context Length / 上下文	131072
Batch / 批处理	✓ true
CPU Cores / 核心数	2
Chip Count / 芯片数	✓ true
Chip Type / 芯片类型	M50

Deployment

Available / 可用	✓ true
Status / 发布状态	Testing
Software Stack / 软件栈版本	2.1.0

Capabilities

Input / 输入	Text
Output / 输出	Text
Features / 特性	gguf w4a8
Inference Engines / 推理引擎	HLIECpp HLIEPython HLIELLama HLIEvLLM
Frameworks / 框架	HLIECpp HLIEPython HLIELLama HLIEvLLM

Description

Description (CN) / 中文描述 Type: Gemma4 | Size: 4 | Quant: w4a8 | Context: 131072 | Chip: M50 | Build: 1.4.0 | Status: Testing

Build Info

Build Version / 编译版本

1.4.0

Timestamps

Created / 创建时间	2026-08-24 00:00:02
Updated / 更新时间	2026-08-24 00:00:02