LLM EXPLORER 59,516 MODELS INDEXED

Midnight Miqu 103B V1.0 EXL2 5.0bpw by Dracones

By Dracones · 5 downloads
HF Repo ↗ Copy URL

Midnight Miqu 103B V1.0 EXL2 5.0bpw is an open-source language model by Dracones. Features: 103b LLM, VRAM: 64.9GB, Context: 31K, Quantized, LLM Explorer Score: 0.11.

  Conversational   Endpoints compatible   Exl2   Llama   Merge   Mergekit   Quantized   Region:us   Safetensors   Sharded   Tensorflow
Model Card on HF 🤗: https://huggingface.co/Dracones/Midnight-Miqu-103B-v1.0_exl2_5.0bpw 

Midnight Miqu 103B V1.0 EXL2 5.0bpw Parameters and Internals

Additional Notes 
This model has been quantized using EXL2 with a precision of 5.0 bits per weight.
LLM NameMidnight Miqu 103B V1.0 EXL2 5.0bpw
Repository 🤗https://huggingface.co/Dracones/Midnight-Miqu-103B-v1.0_exl2_5.0bpw 
Model Size103b
Required VRAM64.9 GB
Updated2026-08-07
MaintainerDracones
Model Typellama
Model Files  8.5 GB: 1-of-8   8.5 GB: 2-of-8   8.6 GB: 3-of-8   8.5 GB: 4-of-8   8.4 GB: 5-of-8   8.6 GB: 6-of-8   8.5 GB: 7-of-8   5.3 GB: 8-of-8
Quantization Typeexl2
Model ArchitectureLlamaForCausalLM
Context Length32764
Model Max Length32764
Transformers Version4.36.2
Tokenizer ClassLlamaTokenizer
Padding Token<unk>
Vocabulary Size32000
Torch Data Typefloat16

Best Alternatives to Midnight Miqu 103B V1.0 EXL2 5.0bpw

Best Alternatives
Context / RAM
Downloads
Likes
... Dark Miqu 103B 5.0bpw H8 EXL231K / 65.1 GB100
...qu 103B V1.5 EXL2 5.0bpw Rpcal31K / 64.9 GB91
... Miqu 103B V1.5 3.5bpw H6 EXL231K / 45.6 GB21
...qu 103B V1.5 EXL2 4.0bpw Rpcal31K / 52.1 GB31
...qu 103B V1.5 EXL2 3.0bpw Rpcal31K / 39.3 GB21
...ght Miqu 103B V1.0 EXL2 4.5bpw31K / 58.7 GB71
...ght Miqu 103B V1.0 EXL2 4.0bpw31K / 52.2 GB61
...ht Miqu 103B V1.0 EXL2 4.25bpw31K / 55.3 GB31
...ght Miqu 103B V1.0 EXL2 3.5bpw31K / 45.7 GB91
...ght Miqu 103B V1.0 EXL2 3.0bpw31K / 39.2 GB42
Note: green Score (e.g. "73.2") means that the model is better than Dracones/Midnight-Miqu-103B-v1.0_exl2_5.0bpw.