Instructions to use mmnga-o/Kimi-K3-REAP50-Width50-UD-gguf with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- llama.cpp
How to use mmnga-o/Kimi-K3-REAP50-Width50-UD-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 mmnga-o/Kimi-K3-REAP50-Width50-UD-gguf:UD-IQ1_M # Run inference directly in the terminal: llama cli -hf mmnga-o/Kimi-K3-REAP50-Width50-UD-gguf:UD-IQ1_M
Install from WinGet (Windows)
winget install llama.cpp # Start a local OpenAI-compatible server with a web UI: llama serve -hf mmnga-o/Kimi-K3-REAP50-Width50-UD-gguf:UD-IQ1_M # Run inference directly in the terminal: llama cli -hf mmnga-o/Kimi-K3-REAP50-Width50-UD-gguf:UD-IQ1_M
Use pre-built binary
# Download pre-built binary from: # https://github.com/ggerganov/llama.cpp/releases # Start a local OpenAI-compatible server with a web UI: ./llama-server -hf mmnga-o/Kimi-K3-REAP50-Width50-UD-gguf:UD-IQ1_M # Run inference directly in the terminal: ./llama-cli -hf mmnga-o/Kimi-K3-REAP50-Width50-UD-gguf:UD-IQ1_M
Build from source code
git clone https://github.com/ggerganov/llama.cpp.git cd llama.cpp cmake -B build cmake --build build -j --target llama-server llama-cli # Start a local OpenAI-compatible server with a web UI: ./build/bin/llama-server -hf mmnga-o/Kimi-K3-REAP50-Width50-UD-gguf:UD-IQ1_M # Run inference directly in the terminal: ./build/bin/llama-cli -hf mmnga-o/Kimi-K3-REAP50-Width50-UD-gguf:UD-IQ1_M
Use Docker
docker model run hf.co/mmnga-o/Kimi-K3-REAP50-Width50-UD-gguf:UD-IQ1_M
- LM Studio
- Jan
- Ollama
How to use mmnga-o/Kimi-K3-REAP50-Width50-UD-gguf with Ollama:
ollama run hf.co/mmnga-o/Kimi-K3-REAP50-Width50-UD-gguf:UD-IQ1_M
- Unsloth Desktop
- Pi
How to use mmnga-o/Kimi-K3-REAP50-Width50-UD-gguf with Pi:
Start the llama.cpp server
# Install llama.cpp: brew install llama.cpp # Start a local OpenAI-compatible server: llama serve -hf mmnga-o/Kimi-K3-REAP50-Width50-UD-gguf:UD-IQ1_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": "mmnga-o/Kimi-K3-REAP50-Width50-UD-gguf:UD-IQ1_M" } ] } } }Run Pi
# Start Pi in your project directory: pi
- Docker Model Runner
How to use mmnga-o/Kimi-K3-REAP50-Width50-UD-gguf with Docker Model Runner:
docker model run hf.co/mmnga-o/Kimi-K3-REAP50-Width50-UD-gguf:UD-IQ1_M
- Lemonade
How to use mmnga-o/Kimi-K3-REAP50-Width50-UD-gguf with Lemonade:
Pull the model
# Download Lemonade from https://lemonade-server.ai/ lemonade pull mmnga-o/Kimi-K3-REAP50-Width50-UD-gguf:UD-IQ1_M
Run and chat with the model
lemonade run user.Kimi-K3-REAP50-Width50-UD-gguf-UD-IQ1_M
List all available models
lemonade list
- Hermes Agent
How to use mmnga-o/Kimi-K3-REAP50-Width50-UD-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 mmnga-o/Kimi-K3-REAP50-Width50-UD-gguf:UD-IQ1_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 mmnga-o/Kimi-K3-REAP50-Width50-UD-gguf:UD-IQ1_M
Run Hermes
hermes
- Atomic Chat
- OpenClaw
How to use mmnga-o/Kimi-K3-REAP50-Width50-UD-gguf with OpenClaw:
Start the llama.cpp server
# Install llama.cpp: brew install llama.cpp # Start a local OpenAI-compatible server: llama serve -hf mmnga-o/Kimi-K3-REAP50-Width50-UD-gguf:UD-IQ1_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 "mmnga-o/Kimi-K3-REAP50-Width50-UD-gguf:UD-IQ1_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"
Add llama.cpp usage instructions
Browse files- README-jp.md +22 -5
- README.md +22 -5
README-jp.md
CHANGED
|
@@ -69,15 +69,32 @@ Width50では、`ffn_gate_exps.weight`、`ffn_up_exps.weight`、`ffn_down_exps.w
|
|
| 69 |
|
| 70 |
使用するフォルダの先頭shardを読み込んでください。
|
| 71 |
|
| 72 |
-
##
|
| 73 |
|
| 74 |
-
Kimi-K3
|
| 75 |
|
| 76 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
| 77 |
|
| 78 |
-
``
|
| 79 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 80 |
```
|
| 81 |
|
|
|
|
|
|
|
| 82 |
強く削減した実験モデルのため、元のKimi-K3とは品質や安定性が異なる場合があります。
|
| 83 |
|
|
|
|
| 69 |
|
| 70 |
使用するフォルダの先頭shardを読み込んでください。
|
| 71 |
|
| 72 |
+
## 使い方
|
| 73 |
|
| 74 |
+
llama.cppの[Kimi-K3 Width対応ブランチ](https://github.com/mmnga/llama.cpp/tree/kimi-k3-width-support)を使用します。
|
| 75 |
|
| 76 |
+
```bash
|
| 77 |
+
git clone --branch kimi-k3-width-support https://github.com/mmnga/llama.cpp
|
| 78 |
+
cmake -S llama.cpp -B llama.cpp/build -DGGML_CUDA=ON -DCMAKE_BUILD_TYPE=Release
|
| 79 |
+
cmake --build llama.cpp/build -j --target llama-server
|
| 80 |
+
```
|
| 81 |
|
| 82 |
+
`UD-IQ1_S`を使う例です。
|
| 83 |
+
|
| 84 |
+
```bash
|
| 85 |
+
./llama.cpp/build/bin/llama-server \
|
| 86 |
+
-m ./UD-IQ1_S/Kimi-K3-UD-IQ1_S-00001-of-00014.gguf \
|
| 87 |
+
-ot ".*.ffn_.*_exps.*=CPU" \
|
| 88 |
+
-ngl 45 \
|
| 89 |
+
--ctx-size 8192 \
|
| 90 |
+
--flash-attn on \
|
| 91 |
+
--jinja \
|
| 92 |
+
--override-kv kimi-k3.expert_shared_count=int:2 \
|
| 93 |
+
--override-kv kimi-k3.expert_shared_feed_forward_length=int:6144 \
|
| 94 |
+
--override-kv kimi-k3.expert_used_count=int:16
|
| 95 |
```
|
| 96 |
|
| 97 |
+
この例はrouted expertsをCPUに置き、32 GBのRTX 5090で動作確認しています。使用可能なVRAMとRAMに合わせて`-ngl`とcontext sizeを調整してください。別の量子化を使う場合も、そのフォルダの先頭shardを指定します。
|
| 98 |
+
|
| 99 |
強く削減した実験モデルのため、元のKimi-K3とは品質や安定性が異なる場合があります。
|
| 100 |
|
README.md
CHANGED
|
@@ -83,15 +83,32 @@ All other tensor data is unchanged. The shared expert metadata is represented as
|
|
| 83 |
|
| 84 |
Load the first shard in the selected folder.
|
| 85 |
|
| 86 |
-
##
|
| 87 |
|
| 88 |
-
|
| 89 |
|
| 90 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
| 91 |
|
| 92 |
-
``
|
| 93 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 94 |
```
|
| 95 |
|
|
|
|
|
|
|
| 96 |
This is a heavily reduced experimental model. Quality and stability may differ from the original Kimi-K3.
|
| 97 |
|
|
|
|
| 83 |
|
| 84 |
Load the first shard in the selected folder.
|
| 85 |
|
| 86 |
+
## Usage
|
| 87 |
|
| 88 |
+
Use the [Kimi-K3 Width support branch](https://github.com/mmnga/llama.cpp/tree/kimi-k3-width-support) of llama.cpp.
|
| 89 |
|
| 90 |
+
```bash
|
| 91 |
+
git clone --branch kimi-k3-width-support https://github.com/mmnga/llama.cpp
|
| 92 |
+
cmake -S llama.cpp -B llama.cpp/build -DGGML_CUDA=ON -DCMAKE_BUILD_TYPE=Release
|
| 93 |
+
cmake --build llama.cpp/build -j --target llama-server
|
| 94 |
+
```
|
| 95 |
|
| 96 |
+
Example using `UD-IQ1_S`:
|
| 97 |
+
|
| 98 |
+
```bash
|
| 99 |
+
./llama.cpp/build/bin/llama-server \
|
| 100 |
+
-m ./UD-IQ1_S/Kimi-K3-UD-IQ1_S-00001-of-00014.gguf \
|
| 101 |
+
-ot ".*.ffn_.*_exps.*=CPU" \
|
| 102 |
+
-ngl 45 \
|
| 103 |
+
--ctx-size 8192 \
|
| 104 |
+
--flash-attn on \
|
| 105 |
+
--jinja \
|
| 106 |
+
--override-kv kimi-k3.expert_shared_count=int:2 \
|
| 107 |
+
--override-kv kimi-k3.expert_shared_feed_forward_length=int:6144 \
|
| 108 |
+
--override-kv kimi-k3.expert_used_count=int:16
|
| 109 |
```
|
| 110 |
|
| 111 |
+
The example was tested with a 32 GB RTX 5090 while keeping routed experts on the CPU. Adjust `-ngl` and context size for your available VRAM and RAM. Load the first shard when using another quantization folder.
|
| 112 |
+
|
| 113 |
This is a heavily reduced experimental model. Quality and stability may differ from the original Kimi-K3.
|
| 114 |
|