Added readme

Author Mitja Felicijan <mitja.felicijan@gmail.com> 2026-02-12 21:12:09 +0100
Committer Mitja Felicijan <mitja.felicijan@gmail.com> 2026-02-12 21:12:09 +0100
Commit 03ebe8c1e276650e63e1db0a97c00d0191f3d520 (patch)
-rw-r--r-- README.md 58
1 files changed, 58 insertions, 0 deletions
diff --git a/README.md b/README.md 
  
 1 
# llmnpc
 
  
 2 
  
 
  
 3 
A command-line LLM inference tool powered by
 
  
 4 
[llama.cpp](https://github.com/ggerganov/llama.cpp) for testing how/if NPC's
 
  
 5 
could use LLM's.
 
  
 6 
  
 
  
 7 
## Building
 
  
 8 
  
 
  
 9 
### Prerequisites
 
  
 10 
  
 
  
 11 
- C compiler (gcc/clang)
 
  
 12 
- CMake
 
  
 13 
- Docker (optional, for containerized use of binaries)
 
  
 14 
  
 
  
 15 
### Build Steps
 
  
 16 
  
 
  
 17 
1. Build llama.cpp libraries:
 
  
 18 
   ```bash
 
  
 19 
   make llamacpp
 
  
 20 
   ```
 
  
 21 
  
 
  
 22 
2. Build the prompt binary:
 
  
 23 
   ```bash
 
  
 24 
   make prompt
 
  
 25 
   ```
 
  
 26 
  
 
  
 27 
## Usage
 
  
 28 
  
 
  
 29 
```bash
 
  
 30 
./prompt -p "Your prompt here"
 
  
 31 
./prompt -m flan-t5-small -p "What is machine learning?"
 
  
 32 
```
 
  
 33 
  
 
  
 34 
### Options
 
  
 35 
  
 
  
 36 
| Flag | Description |
 
  
 37 
|------|-------------|
 
  
 38 
| `-m, --model` | Model to use (default: first model in config) |
 
  
 39 
| `-p, --prompt` | Prompt text (required) |
 
  
 40 
| `-h, --help` | Show help message |
 
  
 41 
  
 
  
 42 
## Models
 
  
 43 
  
 
  
 44 
Configure models in `models.h`. The default model is `flan-t5-small`, expecting a GGUF file at `models/flan-t5-small.F16.gguf`.
 
  
 45 
  
 
  
 46 
## Docker
 
  
 47 
  
 
  
 48 
```bash
 
  
 49 
make docker
 
  
 50 
```
 
  
 51 
  
 
  
 52 
This builds a Docker image and drops you into a shell with the prompt binary and models available at `/app/`.
 
  
 53 
  
 
  
 54 
## Cleaning
 
  
 55 
  
 
  
 56 
```bash
 
  
 57 
make clean
 
  
 58 
```