LLM EXPLORER
 LLM Explorer: A Curated Large Language Model Directory and Analytics
Dark Theme
  Agents   Benchmarks   LLM  List    LLM Hosting    LLM Leaderboards    Blog    Newsfeed    Advertise

Midnight Miqu 103B V1.5 3.5bpw H6 EXL2 by LoneStriker

 »  All LLMs  »  LoneStriker  »  Midnight Miqu 103B V1.5 3.5bpw H6 EXL2   URL Share it on

  3   

Midnight Miqu 103B V1.5 3.5bpw H6 EXL2 is an open-source language model by LoneStriker. Features: 103b LLM, VRAM: 45.6GB, Context: 31K, Quantized, Merged, LLM Explorer Score: 0.12.

  Merged Model   Conversational   Endpoints compatible   Exl2   Llama   Quantized   Region:us   Safetensors   Sharded   Tensorflow
Model Card on HF 🤗: https://huggingface.co/LoneStriker/Midnight-Miqu-103B-v1.5-3.5bpw-h6-exl2 

Midnight Miqu 103B V1.5 3.5bpw H6 EXL2 Benchmarks

LLME Score: 0.12148
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
Midnight Miqu 103B V1.5 3.5bpw H6 EXL2 (LoneStriker/Midnight-Miqu-103B-v1.5-3.5bpw-h6-exl2)
🌟 Advertise your project 🚀

Midnight Miqu 103B V1.5 3.5bpw H6 EXL2 Parameters and Internals

Model Type 
text generation
Additional Notes 
This is a 103B frankenmerge of Midnight-Miqu-70B-v1.5 with itself. Model is capable of 32K context.
LLM NameMidnight Miqu 103B V1.5 3.5bpw H6 EXL2
Repository 🤗https://huggingface.co/LoneStriker/Midnight-Miqu-103B-v1.5-3.5bpw-h6-exl2 
Merged ModelYes
Model Size103b
Required VRAM45.6 GB
Updated2026-04-21
MaintainerLoneStriker
Model Typellama
Model Files  8.5 GB: 1-of-6   8.6 GB: 2-of-6   8.5 GB: 3-of-6   8.5 GB: 4-of-6   8.5 GB: 5-of-6   3.0 GB: 6-of-6
Quantization Typeexl2
Model ArchitectureLlamaForCausalLM
Context Length32764
Model Max Length32764
Transformers Version4.35.2
Tokenizer ClassLlamaTokenizer
Padding Token<unk>
Vocabulary Size32000
Torch Data Typefloat16

Best Alternatives to Midnight Miqu 103B V1.5 3.5bpw H6 EXL2

Best Alternatives
Context / RAM
Downloads
Likes
... Dark Miqu 103B 5.0bpw H8 EXL231K / 65.1 GB50
...qu 103B V1.5 EXL2 5.0bpw Rpcal31K / 64.9 GB51
...qu 103B V1.5 EXL2 3.0bpw Rpcal31K / 39.3 GB41
...qu 103B V1.5 EXL2 4.0bpw Rpcal31K / 52.1 GB41
...ght Miqu 103B V1.0 EXL2 4.5bpw31K / 58.7 GB51
...ght Miqu 103B V1.0 EXL2 4.0bpw31K / 52.2 GB11
...ht Miqu 103B V1.0 EXL2 4.25bpw31K / 55.3 GB11
...ght Miqu 103B V1.0 EXL2 3.0bpw31K / 39.2 GB22
...ght Miqu 103B V1.0 EXL2 2.4bpw31K / 31.6 GB31
...ght Miqu 103B V1.0 EXL2 3.5bpw31K / 45.7 GB31
Note: green Score (e.g. "73.2") means that the model is better than LoneStriker/Midnight-Miqu-103B-v1.5-3.5bpw-h6-exl2.

Rank the Midnight Miqu 103B V1.5 3.5bpw H6 EXL2 Capabilities

🆘 Have you tried this model? Rate its performance. This feedback would greatly assist ML community in identifying the most suitable model for their needs. Your contribution really does make a difference! 🌟

Instruction Following and Task Automation  
Factuality and Completeness of Knowledge  
Censorship and Alignment  
Data Analysis and Insight Generation  
Text Generation  
Text Summarization and Feature Extraction  
Code Generation  
Multi-Language Support and Translation  

What open-source LLMs or SLMs are you in search of? 53999 in total.

Email us: info@extractum.io. Our Privacy Policy  | Terms and Conditions  | Suggest an improvement.
Our Social Media →  
Original data from HuggingFace, OpenCompass and various public git repos.
Check out Ag3ntum — our secure, self-hosted AI agent for server management.
Release v20260328a
  Support LLM Explorer