Platform
GA
Open source
llama.cpp
Running GGUF anywhere, including CPU/Apple Silicon
Release2023-03
Context—
Max output—
ParametersUndisclosed
Local / VRAM2–24 GB
LicenseOpen source
Overview
llama.cpp is a model in the Platform category from ggml-org. It was released in 2023-03. The vendor is based in US. Best for: Running GGUF anywhere, including CPU/Apple Silicon.
Strengths
- Wide quantization
- CPU/Metal/CUDA
- Portable
Limits
- Slow on large models
- No UI
Specifications
| Vendor | ggml-org |
|---|---|
| Category | Platform |
| Release | 2023-03 |
| Status | GA |
| License | Open source |
| License name | MIT |
| Commercial use | Allowed |
| Input modalities | Text |
| Output modalities | Text |
| Prompt types | Text |
| Negative prompts | No |
| Max outputs | 1 |
| Local deployment | Yes |
| VRAM | 2–24 GB |
| System RAM | 4–64 GB |
| API access | Yes |
| Pricing | Free |
| Confidence | High |
| Vendor country | US |
| Vendor type | Platform |
Pricing & hardware
Free
This model can be self-hosted or run locally. Approximate GPU memory: 2–24 GB. Approximate system RAM: 4–64 GB. API access is available.