How to use from
llama.cppInstall from WinGet (Windows)
winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf Abiray/BitCPM4-CANN-1B-GGUF:# Run inference directly in the terminal:
llama cli -hf Abiray/BitCPM4-CANN-1B-GGUF: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 Abiray/BitCPM4-CANN-1B-GGUF:# Run inference directly in the terminal:
./llama-cli -hf Abiray/BitCPM4-CANN-1B-GGUF: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 Abiray/BitCPM4-CANN-1B-GGUF:# Run inference directly in the terminal:
./build/bin/llama-cli -hf Abiray/BitCPM4-CANN-1B-GGUF:Use Docker
docker model run hf.co/Abiray/BitCPM4-CANN-1B-GGUF:Quick Links
Abiray/BitCPM4-CANN-1B-GGUF
This repository contains quantized GGUF formats of the openbmb/BitCPM4-CANN-1B model, heavily optimized for local inference using llama.cpp, text-generation-webui, LM Studio, Ollama, and other compatible backend frameworks.
Model Information
- Original Model: openbmb/BitCPM4-CANN-1B
- Architecture: BitCPM (1 Billion Parameters)
Available Files & Hardware Compatibility
The following quantization formats are available. Because this is a 1-Billion parameter model, it is highly efficient and can easily run on consumer CPUs, ultra-low-end hardware, and mobile devices.
| Filename | Quant Type | File Size | Description |
|---|---|---|---|
| BitCPM4-CANN-1B-Q8_0.gguf | 8-bit | 1.73 GB | Highest quality, almost indistinguishable from the unquantized base model. Very fast inference. |
| BitCPM4-CANN-1B-Q6_K.gguf | 6-bit | 1.33 GB | Excellent quality with near-zero noticeable degradation. Highly recommended. |
| BitCPM4-CANN-1B-Q5_K_M.gguf | 5-bit | 1.16 GB | Great balance of file size, text generation speed, and logic retention. |
| BitCPM4-CANN-1B-Q5_K_S.gguf | 5-bit | 1.14 GB | Minor variant of Q5_K_M optimized slightly more for size. |
| BitCPM4-CANN-1B-Q4_K_M.gguf | 4-bit | 1.00 GB | Recommended. The ideal sweet-spot for 4-bit formats, striking an incredible performance-to-size ratio. |
| BitCPM4-CANN-1B-Q4_K_S.gguf | 4-bit | 958 MB | Extremely small and fast. Drops below the 1GB mark, making it perfect for lightweight deployments. |
| BitCPM4-CANN-1B-Q3_K_M.gguf | 3-bit | 824 MB | Maximum compression. Use only if working under severe memory bottlenecks. |
How to Run
Using llama.cpp (Command Line)
If you have compiled llama.cpp, you can run the model directly from your terminal. Replace the filename with the specific version you downloaded:
./llama-cli \
-m BitCPM4-CANN-1B-Q4_K_M.gguf \
-p "Explain the concept of artificial intelligence to a five-year-old." \
-n 256 \
-c 2048 \
--temp 0.7
- Downloads last month
- 51
Hardware compatibility
Log In to add your hardware
3-bit
4-bit
5-bit
6-bit
8-bit
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support
Model tree for Abiray/BitCPM4-CANN-1B-GGUF
Base model
openbmb/BitCPM-CANN-1B
Install (macOS, Linux)
# Start a local OpenAI-compatible server with a web UI: llama serve -hf Abiray/BitCPM4-CANN-1B-GGUF:# Run inference directly in the terminal: llama cli -hf Abiray/BitCPM4-CANN-1B-GGUF: