---
language:
  - en
license: apache-2.0
tags:
  - code
  - local
  - private
  - ollama
  - gguf
  - venti
base_model: Qwen/Qwen2.5-Coder-7B
pipeline_tag: text-generation
---

# Venti MiseAI 1.1

**Intelligence that lives on your machine.**

MiseAI is a powerful, private AI assistant built by [Venti Labs](https://venti-labs.xyz). It runs 100% locally on your hardware — no cloud, no API keys, no data leaving your device.

## Highlights

- 🧠 **7B Parameters** — Fine-tuned from Qwen 2.5 Coder 7B
- 🔒 **Fully Private** — Runs offline, no internet required after download
- 💻 **Expert Coder** — Production-ready code generation and refactoring
- ⚡ **8GB VRAM** — Optimized to run on consumer GPUs
- 📦 **GGUF Format** — Ready for Ollama, llama.cpp, LM Studio

## Quick Start (Ollama)

```bash
ollama run ventilabs/miseai
```

Or install the Venti CLI:

```powershell
irm venti-labs.xyz/install | iex
venti launch mise
```

## Model Details

| Property | Value |
|---|---|
| Base Model | Qwen 2.5 Coder 7B |
| Fine-tuning | LoRA (QLoRA) |
| Quantization | Q8_0 |
| File Size | ~8.1 GB |
| Context Window | 16,384 tokens |
| Max Output | 8,192 tokens |

## Use Cases

- **Code Generation** — Write production-ready code in any language
- **Code Refactoring** — Optimize and restructure existing codebases
- **Problem Solving** — Step-by-step reasoning through complex challenges
- **Technical Writing** — Documentation, README files, and technical articles

## Links

- 🌐 Website: [venti-labs.xyz](https://venti-labs.xyz)
- 🐙 Ollama: [ventilabs/miseai](https://ollama.com/ventilabs/miseai)

---

Built with ❤️ by Venti Labs © 2026