Instructions to use webmp3/Sakura-Nanbeige4.3-33B-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use webmp3/Sakura-Nanbeige4.3-33B-GGUF with llama.cpp:

Install (macOS, Linux)

curl -LsSf https://llama.app/install.sh | sh
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf webmp3/Sakura-Nanbeige4.3-33B-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf webmp3/Sakura-Nanbeige4.3-33B-GGUF:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf webmp3/Sakura-Nanbeige4.3-33B-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf webmp3/Sakura-Nanbeige4.3-33B-GGUF:Q4_K_M

Build from source code

git clone https://github.com/ggml-org/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-app
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama serve -hf webmp3/Sakura-Nanbeige4.3-33B-GGUF:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama cli -hf webmp3/Sakura-Nanbeige4.3-33B-GGUF:Q4_K_M

Use Docker

docker model run hf.co/webmp3/Sakura-Nanbeige4.3-33B-GGUF:Q4_K_M

Llama new
LM Studio
Jan
Ollama
How to use webmp3/Sakura-Nanbeige4.3-33B-GGUF with Ollama:
```
ollama run hf.co/webmp3/Sakura-Nanbeige4.3-33B-GGUF:Q4_K_M
```
Unsloth Desktop

How to use webmp3/Sakura-Nanbeige4.3-33B-GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf webmp3/Sakura-Nanbeige4.3-33B-GGUF:Q4_K_M

Configure the model in Pi

# Install Pi:
npm install -g @earendil-works/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "llama-cpp": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "webmp3/Sakura-Nanbeige4.3-33B-GGUF:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use webmp3/Sakura-Nanbeige4.3-33B-GGUF with Docker Model Runner:
```
docker model run hf.co/webmp3/Sakura-Nanbeige4.3-33B-GGUF:Q4_K_M
```

Lemonade

How to use webmp3/Sakura-Nanbeige4.3-33B-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull webmp3/Sakura-Nanbeige4.3-33B-GGUF:Q4_K_M

Run and chat with the model

lemonade run user.Sakura-Nanbeige4.3-33B-GGUF-Q4_K_M

List all available models

lemonade list

Hermes Agent

How to use webmp3/Sakura-Nanbeige4.3-33B-GGUF with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf webmp3/Sakura-Nanbeige4.3-33B-GGUF:Q4_K_M

Configure Hermes

# Install Hermes:
curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash
hermes setup
# Point Hermes at the local server:
hermes config set model.provider custom
hermes config set model.base_url http://127.0.0.1:8080/v1
hermes config set model.default webmp3/Sakura-Nanbeige4.3-33B-GGUF:Q4_K_M

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use webmp3/Sakura-Nanbeige4.3-33B-GGUF with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf webmp3/Sakura-Nanbeige4.3-33B-GGUF:Q4_K_M

Configure OpenClaw

# Install OpenClaw:
npm install -g openclaw@latest
# Register the local server and set it as the default model:
openclaw onboard --non-interactive --mode local \
  --auth-choice custom-api-key \
  --custom-base-url http://127.0.0.1:8080/v1 \
  --custom-model-id "webmp3/Sakura-Nanbeige4.3-33B-GGUF:Q4_K_M" \
  --custom-provider-id llama-cpp \
  --custom-compatibility openai \
  --custom-text-input \
  --accept-risk \
  --skip-health

Run OpenClaw

openclaw agent --local --agent main --message "Hello from Hugging Face"

Sakura · GGUF quantization of Nanbeige/Nanbeige4.3-33B

Independent community quantization. Not an official release of the upstream author.

1. Model Overview

Architecture: llama · 33.1 B parameters · source formats: safetensors, bfloat16.

Original: 66.2 GB (BF16 GGUF, the source of these quantizations).

Quant	Size	Compared with the original
Q2_K	12.7 GB	81 % smaller
Q3_K_M	16.3 GB	75 % smaller
Q4_K_M	20.1 GB	70 % smaller
Q5_K_M	23.5 GB	65 % smaller

2. Original Model / Attribution

Original model: Nanbeige/Nanbeige4.3-33B, commit c1093c501cc6460427c0ac634db7223d0450c0ee (model card at that commit)
Declared base models: Nanbeige/Nanbeige4.3-33B-Base (relation: finetune)
Quantized from: bartowski/Nanbeige_Nanbeige4.3-33B-GGUF at 9beb68421fdf7be095093a6888325fa2c83ddd38, file Nanbeige_Nanbeige4.3-33B-bf16.gguf (SHA-256 09da2d7593011d29c2b3aa139bfdc0b32b38f21ec58f96f2c43ad083a2d2c068; provenance verified: True)
Quantized by webmp3. Upstream authors retain their license and attribution.

3. Files

File	Bytes	SHA-256	Quant
`Sakura-Nanbeige4.3-33B-Q2_K.gguf`	12657044352	`8d865ed7664ff9e359d70a4aa834314c2f16f50fb008a28fabfe87d304488b2e`	Q2_K
`Sakura-Nanbeige4.3-33B-Q3_K_M.gguf`	16261286784	`18537b39533caa15d65a775200e79145b326810e3efa250a19a57e2a7bd4a833`	Q3_K_M
`Sakura-Nanbeige4.3-33B-Q4_K_M.gguf`	20051495808	`62bc6475c38cf914dc6c0c53fda51d0d01053b4642d6278701f46dcd7542b2b2`	Q4_K_M
`Sakura-Nanbeige4.3-33B-Q5_K_M.gguf`	23504980864	`a76fae2099246d5c4429962849bad35298b8d25820dd201629f2de0b05d8a14d`	Q5_K_M

4. How These Were Made

This route performs existing-GGUF requantization from the pinned BF16 source GGUF named above. This job did not convert the upstream BF16/safetensors checkpoint.

No importance-matrix calibration was run by this job.

The quant label names the llama.cpp preset. Stock llama.cpp keeps some tensors (for example normalization weights) in higher precision, so not every tensor is stored in the named type; the table shows what each file really contains.

File	Tensors by type
`Sakura-Nanbeige4.3-33B-Q2_K.gguf`	F32: 105, Q2_K: 209, Q3_K: 104, Q4_K: 52, Q6_K: 1
`Sakura-Nanbeige4.3-33B-Q3_K_M.gguf`	F32: 105, Q3_K: 209, Q4_K: 151, Q5_K: 5, Q6_K: 1
`Sakura-Nanbeige4.3-33B-Q4_K_M.gguf`	F32: 105, Q4_K: 313, Q6_K: 53
`Sakura-Nanbeige4.3-33B-Q5_K_M.gguf`	F32: 105, Q5_K: 313, Q6_K: 53

5. Verification

GGUF structure valid: True
Real inference test (every file loaded by llama.cpp and used to generate text once): yes

File	llama.cpp build	Executable SHA-256
`Sakura-Nanbeige4.3-33B-Q2_K.gguf`	`b11669-620d2225c`	`3b7167843ffb0f3677be964dc80cfc59a0167c82e11eff62f6334f66e831f209`
`Sakura-Nanbeige4.3-33B-Q3_K_M.gguf`	`b11669-620d2225c`	`3b7167843ffb0f3677be964dc80cfc59a0167c82e11eff62f6334f66e831f209`
`Sakura-Nanbeige4.3-33B-Q4_K_M.gguf`	`b11669-620d2225c`	`3b7167843ffb0f3677be964dc80cfc59a0167c82e11eff62f6334f66e831f209`
`Sakura-Nanbeige4.3-33B-Q5_K_M.gguf`	`b11669-620d2225c`	`3b7167843ffb0f3677be964dc80cfc59a0167c82e11eff62f6334f66e831f209`

6. Download and Usage

llama-cli -m <file>.gguf -p "Hello" -n 32

Needs a llama.cpp build that supports this architecture. If a quantization is split into numbered parts, keep them together and load the first one.

7. Limitations

No coding benchmark, perplexity or quality score is claimed unless a measured value is listed above.

8. License Information

Source license: apache-2.0. apache-2.0 allows redistribution with attribution

中文说明 · 樱花 Sakura（点击展开）

1. 概述

独立的社区量化版本，并非上游作者的官方发布。

架构：llama · 33.1 B 参数 · 源格式：safetensors, bfloat16。

原始模型：66.2 GB（BF16 GGUF，这些量化版本的来源）。

量化	大小	相比原始模型
Q2_K	12.7 GB	小 81 %
Q3_K_M	16.3 GB	小 75 %
Q4_K_M	20.1 GB	小 70 %
Q5_K_M	23.5 GB	小 65 %

2. 原始模型与署名

原始模型：Nanbeige/Nanbeige4.3-33B，提交 c1093c501cc6460427c0ac634db7223d0450c0ee（该提交的模型卡）
声明的基础模型：Nanbeige/Nanbeige4.3-33B-Base（关系：finetune）
量化来源：bartowski/Nanbeige_Nanbeige4.3-33B-GGUF，提交 9beb68421fdf7be095093a6888325fa2c83ddd38，文件 Nanbeige_Nanbeige4.3-33B-bf16.gguf（SHA-256 09da2d7593011d29c2b3aa139bfdc0b32b38f21ec58f96f2c43ad083a2d2c068；出处已验证：True）
量化者：webmp3。上游作者保留其许可证与署名。

3. 文件

文件	字节数	SHA-256	量化
`Sakura-Nanbeige4.3-33B-Q2_K.gguf`	12657044352	`8d865ed7664ff9e359d70a4aa834314c2f16f50fb008a28fabfe87d304488b2e`	Q2_K
`Sakura-Nanbeige4.3-33B-Q3_K_M.gguf`	16261286784	`18537b39533caa15d65a775200e79145b326810e3efa250a19a57e2a7bd4a833`	Q3_K_M
`Sakura-Nanbeige4.3-33B-Q4_K_M.gguf`	20051495808	`62bc6475c38cf914dc6c0c53fda51d0d01053b4642d6278701f46dcd7542b2b2`	Q4_K_M
`Sakura-Nanbeige4.3-33B-Q5_K_M.gguf`	23504980864	`a76fae2099246d5c4429962849bad35298b8d25820dd201629f2de0b05d8a14d`	Q5_K_M

4. 制作方式

该路线从上方标明的固定版本 BF16 源 GGUF 进行现有 GGUF 的再量化。本任务没有转换上游的 BF16/safetensors 检查点。

本任务没有运行重要性矩阵（imatrix）校准。

量化标签指 llama.cpp 的预设名称。标准 llama.cpp 会把部分张量（例如归一化权重）保留为更高精度，因此并非每个张量都以所标注的类型存储；下表列出每个文件实际包含的内容。

文件	按类型的张量数
`Sakura-Nanbeige4.3-33B-Q2_K.gguf`	F32: 105, Q2_K: 209, Q3_K: 104, Q4_K: 52, Q6_K: 1
`Sakura-Nanbeige4.3-33B-Q3_K_M.gguf`	F32: 105, Q3_K: 209, Q4_K: 151, Q5_K: 5, Q6_K: 1
`Sakura-Nanbeige4.3-33B-Q4_K_M.gguf`	F32: 105, Q4_K: 313, Q6_K: 53
`Sakura-Nanbeige4.3-33B-Q5_K_M.gguf`	F32: 105, Q5_K: 313, Q6_K: 53

5. 验证

GGUF 结构有效：True
真实推理测试（每个文件都由 llama.cpp 加载并生成一次文本）：yes

文件	llama.cpp 版本	可执行文件 SHA-256
`Sakura-Nanbeige4.3-33B-Q2_K.gguf`	`b11669-620d2225c`	`3b7167843ffb0f3677be964dc80cfc59a0167c82e11eff62f6334f66e831f209`
`Sakura-Nanbeige4.3-33B-Q3_K_M.gguf`	`b11669-620d2225c`	`3b7167843ffb0f3677be964dc80cfc59a0167c82e11eff62f6334f66e831f209`
`Sakura-Nanbeige4.3-33B-Q4_K_M.gguf`	`b11669-620d2225c`	`3b7167843ffb0f3677be964dc80cfc59a0167c82e11eff62f6334f66e831f209`
`Sakura-Nanbeige4.3-33B-Q5_K_M.gguf`	`b11669-620d2225c`	`3b7167843ffb0f3677be964dc80cfc59a0167c82e11eff62f6334f66e831f209`

6. 下载与使用

llama-cli -m <file>.gguf -p "Hello" -n 32

需要支持该架构的 llama.cpp 版本。如果某个量化被拆分为带编号的多个部分，请将它们放在一起并加载第一个。

7. 局限

除非上文列出了实测数值，否则不声称任何编程基准、困惑度或质量分数。

8. 许可证信息

源许可证：apache-2.0。apache-2.0 allows redistribution with attribution

完整独立的中文版：README_zh.md

Downloads last month: 148

GGUF

Model size

33B params

Architecture

llama

Hardware compatibility

2-bit

3-bit

4-bit

5-bit

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for webmp3/Sakura-Nanbeige4.3-33B-GGUF

Base model

Nanbeige/Nanbeige4.3-33B-Base

Finetuned

Nanbeige/Nanbeige4.3-33B

Quantized

(94)

this model