Midnight-Miqu-103B-v1.0 Open Source Model - An Efficient Hybrid Model Supporting 32K Context Length

Home

Midnight Miqu 103B V1.0

Developed by sophosympatheia

A 103B parameter hybrid model based on the leaked Miqu model, supporting 32K context length

Large Language Model

Transformers

Open Source License:Other #32K long text processing #Large language model #Non-commercial use

Downloads 18

Release Time : 3/2/2024

Model Overview

This is a 103B hybrid model combining Midnight-Miqu-70B-v1.0 with itself, based on the Miqu architecture, supporting long text processing

Model Features

Long context support

Supports 32K tokens context length, suitable for processing long documents

Large parameter scale

103B parameter hybrid model providing powerful text generation capabilities

Multiple quantization versions

Offers various quantization versions (GGUF/EXL2) to accommodate different hardware needs

Model Capabilities

Long text generation

Multi-turn dialogue

Text understanding and analysis

Use Cases

Text processing

Long document summarization

Process documents up to 32K tokens and generate summaries

Complex dialogue systems

Supports multi-turn complex dialogues while maintaining long-context consistency

🚀 MidnightMiqu 103B Frankenmerge

This is a 103B frankenmerge model, based on Miqu and capable of 32K context.

🚀 Quick Start

This is a 103B frankenmerge of sophosympatheia/Midnight-Miqu-70B-v1.0 with itself. Please see that model card for details and usage instructions. This model is based on Miqu so it's capable of 32K context.

💸 Donations

If you feel like saying thanks with a donation, I'm on Ko-Fi

💾 Quantizations

GGUF
- Dracones/Midnight-Miqu-103B-v1.0-GGUF
EXL2
If you don't see something you're looking for, try searching Hugging Face. There may be newer quants available than what I've documented here.

📄 License

⚠️ Important Note

152334H/miqu-1-70b-sf was based on a leaked version of one of Mistral's models. All miqu-derived models, including this merge, are only suitable for personal use. Mistral has been cool about it so far, but you should be aware that by downloading this merge you are assuming whatever legal risk is iherent in acquiring and using a model based on leaked weights. This merge comes with no warranties or guarantees of any kind, but you probably already knew that. I am not a lawyer and I do not profess to know what we have gotten ourselves into here. You should consult with a lawyer before using any Hugging Face model beyond private use... but definitely don't use this one for that!

🛠️ Merge Details

🔀 Merge Method

This model was merged using the passthrough merge method.

🧩 Models Merged

The following models were included in the merge:

sophosympatheia/Midnight-Miqu-70B-v1.0

⚙️ Configuration

The following YAML configuration was used to produce this model:

slices:
  - sources:
      - model: /home/llm/mergequant/models/midnight-miqu-70b
        layer_range: [0, 40] # 40
  - sources:
      - model: /home/llm/mergequant/models/midnight-miqu-70b
        layer_range: [20, 60] # 40
  - sources:
      - model: /home/llm/mergequant/models/midnight-miqu-70b
        layer_range: [40, 80] # 40
merge_method: passthrough
dtype: float16

Property	Details
Base Model	None
Library Name	transformers
Tags	mergekit, merge
License	other

Featured Recommended AI Models

Empowering the Future, Your AI Solution Knowledge Base

English 简体中文繁體中文にほんご