<p align="center">
<img src="assets/img/logo.png" alt="aha logo" width="120"/>
</p>
<p align="center">
<a href="https://github.com/jhqxxx/aha/stargazers">
<img src="https://img.shields.io/github/stars/jhqxxx/aha" alt="GitHub Stars">
</a>
<a href="https://github.com/jhqxxx/aha/issues">
<img src="https://img.shields.io/github/issues/jhqxxx/aha" alt="GitHub Issues">
</a>
<a href="https://github.com/jhqxxx/aha/blob/main/LICENSE">
<img src="https://img.shields.io/github/license/jhqxxx/aha" alt="GitHub License">
</a>
</p>
<p align="center">
<a href="README.zh-CN.md">简体中文</a> | <strong>English</strong>
</p>
# aha
**Lightweight AI Inference Engine — All-in-one Solution for Text, Vision, Speech, and OCR**
aha is a high-performance, cross-platform AI inference engine built with Rust and the Candle framework. It brings state-of-the-art AI models to your local machine—no API keys, no cloud dependencies, just pure, fast AI running directly on your hardware.
## Changelog
### v0.2.4 (2026-03-23)
- add LFM2.5-1.2B-Instruct
- add LFM2-1.2B
### v0.2.3 (2026-03-18)
- add DeepSeek-OCR-2
### 2026-03-17
- add PaddleOCR-VL1.5 model
- fix qwen3.5 position_ids create bug
- cli param add
- gguf_path: Local GGUF model weight path (required for loading models with GGUF)
- mmproj_path: Local path to mmproj GGUF weights (required for multimodal GGUF loading)
- WhichModel add qwen3.5-gguf
### 2026-03-16
- Added Qwen3.5 mmproj
### 2026-03-14
- update rust version
- Added Qwen3.5 gguf support, but the 4B model still has issues; to be resolved.
### v0.2.2 (2026-03-07)
- Added GLM-OCR model
### v0.2.1 (2026-03-05)
- Added Qwen3.5 model
**[View full changelog](docs/changelog.md)** →
## Quick Start
### Installation
```bash
git clone https://github.com/jhqxxx/aha.git
cd aha
cargo build --release
```
**Optional Features:**
```bash
# CUDA (NVIDIA GPU acceleration)
cargo build --release --features cuda
# Metal (Apple GPU acceleration for macOS)
cargo build --release --features metal
# Flash Attention (faster inference)
cargo build --release --features cuda,flash-attn
# FFmpeg (multimedia processing)
cargo build --release --features ffmpeg
```
### CLI Quick Reference
```bash
# List all supported models
aha list
# Download model only
aha download -m qwen3asr-0.6b
# Download model and start service
aha -m qwen3asr-0.6b
# Run inference directly (without starting service)
aha run -m qwen3asr-0.6b -i "audio.wav"
# Start service only (model already downloaded)
aha serv -m qwen3asr-0.6b -p 10100
```
### Chat
```bash
aha serv -m qwen3-0.6b -p 10100
```
Then use the unified (OpenAI-compatible) API:
```bash
curl http://localhost:10100/chat/completions \
-H "Content-Type: application/json" \
-d '{
"model": "qwen3-0.6b",
"messages": [{"role": "user", "content": "Hello!"}],
"stream": false
}
'
```
### Supported Models
| **Text** | Qwen3, MiniCPM4, <br> LFM2-1.2B, LFM2.5-1.2B-Instruct |
| **Vision** | Qwen2.5-VL, Qwen3-VL, Qwen3.5 |
| **OCR** | DeepSeek-OCR, DeepSeek-OCR-2 , <br> Hunyuan-OCR, PaddleOCR-VL, PaddleOCR-VL1.5 |
| **ASR** | GLM-ASR-Nano, Fun-ASR-Nano, Qwen3-ASR |
| **Audio** | VoxCPM, VoxCPM1.5 |
| **Image** | RMBG-2.0 (background removal) |
## Documentation
| [Getting Started](docs/getting-started.md) | First steps with aha |
| [Installation](docs/installation.md) | Detailed installation guide |
| [CLI Reference](docs/cli.md) | Command-line interface |
| [API Documentation](docs/api.md) | Library & REST API |
| [Supported Models](docs/supported-models.md) | Available AI models |
| [Concepts](docs/concepts.md) | Architecture & design |
| [Development](docs/development.md) | Contributing guide |
| [Changelog](docs/changelog.md) | Version history |
## Why aha?
- **🚀 High-Performance Inference** - Powered by Candle framework for efficient tensor computation and model inference
- **🔧 Unified Interface** — One tool for text, vision, speech, and OCR
- **📦 Local-First** — All processing runs locally, no data leaves your machine
- **🎯 Cross-Platform** — Works on Linux, macOS, and Windows
- **⚡ GPU Accelerated** — Optional CUDA support for faster inference
- **🛡️ Memory Safe** — Built with Rust for reliability
- **🧠 Attention Optimization** - Optional Flash Attention support for optimized long sequence processing
## Development
### Using aha as a Library
> cargo add aha
```rust
# VoxCPM example
use aha::models::voxcpm::generate::VoxCPMGenerate;
use aha::utils::audio_utils::save_wav;
use anyhow::Result;
fn main() -> Result<()> {
let model_path = "xxx/openbmb/VoxCPM-0.5B/";
let mut voxcpm_generate = VoxCPMGenerate::init(model_path, None, None)?;
let generate = voxcpm_generate.generate(
"The sun is shining bright, flowers smile at me, birds say early early early".to_string(),
None,
None,
2,
100,
10,
2.0,
false,
6.0,
)?;
let _ = save_wav(&generate, "voxcpm.wav")?;
Ok(())
}
```
### Extending New Models
- Create new model file in src/models/
- Export in src/models/mod.rs
- Add support for CLI model inference in src/exec/
- Add tests and examples in tests/
## Features
- High-performance inference via Candle framework
- Multi-modal model support (vision, language, speech)
- Clean, easy-to-use API design
- Minimal dependencies, compact binaries
- Flash Attention support for long sequences
- FFmpeg support for multimedia processing
## License
Apache-2.0 — See [LICENSE](LICENSE) for details.
## Acknowledgments
- [Candle](https://github.com/huggingface/candle) - Excellent Rust ML framework
- All model authors and contributors
## Wechat
---
<p align="center">
<sub>Built with ❤️ by the aha team</sub>
</p>
<p align="center">
<sub>We're continuously expanding our model support. Contributions are welcome!</sub>
</p>
<p align="center">
<sub>If this project helps you, please consider giving us a ⭐ Star!</sub>
</p>