Midnight Miqu 70B V1.0 by sophosympatheia

 ยป  All LLMs  ยป  sophosympatheia  ยป  Midnight Miqu 70B V1.0   URL Share it on

  Merged Model   Arxiv:2307.11760   Autotrain compatible   Conversational   Endpoints compatible   Llama   Region:us   Safetensors   Sharded   Tensorflow

Midnight Miqu 70B V1.0 Benchmarks

nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
Midnight Miqu 70B V1.0 (sophosympatheia/Midnight-Miqu-70B-v1.0)

Midnight Miqu 70B V1.0 Parameters and Internals

Model Type 
roleplaying, storytelling
Use Cases 
Primary Use Cases:
roleplaying, storytelling
Limitations:
Only suitable for personal use
Considerations:
Provides succinct, coherent, and vivid accounts based on instructions.
Additional Notes 
This is a SLERP merge between 152334H/miqu-1-70b-sf and sophosympatheia/Midnight-Rose-70B-v2.0.3, designed using mergekit.
Training Details 
Methodology:
SLERP merge using mergekit
Input Output 
Performance Tips:
You can run this model out to 32K context with alpha_rope set to 1. Limited testing shows coherence out to 64K using alpha_rope 2.5. Try different sampling settings for creative work.
LLM NameMidnight Miqu 70B V1.0
Repository ๐Ÿค—https://huggingface.co/sophosympatheia/Midnight-Miqu-70B-v1.0 
Merged ModelYes
Model Size70b
Required VRAM139 GB
Updated2025-03-18
Maintainersophosympatheia
Model Typellama
Model Files  10.1 GB: 1-of-15   9.8 GB: 2-of-15   9.6 GB: 3-of-15   9.8 GB: 4-of-15   9.9 GB: 5-of-15   9.7 GB: 6-of-15   10.0 GB: 7-of-15   9.9 GB: 8-of-15   9.7 GB: 9-of-15   9.6 GB: 10-of-15   10.0 GB: 11-of-15   9.8 GB: 12-of-15   9.6 GB: 13-of-15   10.4 GB: 14-of-15   1.1 GB: 15-of-15
Model ArchitectureLlamaForCausalLM
Licenseother
Context Length32764
Model Max Length32764
Transformers Version4.36.2
Tokenizer ClassLlamaTokenizer
Padding Token<unk>
Vocabulary Size32000
Torch Data Typefloat16

Quantized Models of the Midnight Miqu 70B V1.0

Model
Likes
Downloads
VRAM
Midnight Miqu 70B V1.0 GGUF425129 GB

Best Alternatives to Midnight Miqu 70B V1.0

Best Alternatives
Context / RAM
Downloads
Likes
... Chat 1048K Chinese Llama3 70B1024K / 141.9 GB40355
... 3 70B Instruct Gradient 1048K1024K / 141.9 GB108121
... Chat 1048K Chinese Llama3 70B1024K / 141.9 GB2305
Llama3 Function Calling 1048K1024K / 141.9 GB81
...a 3 70B Instruct Gradient 524K512K / 141.9 GB5223
...a 3 70B Instruct Gradient 262K256K / 141.9 GB7655
...ama 3 70B Arimas Story RP V2.0256K / 141.1 GB193
...ama 3 70B Arimas Story RP V1.5256K / 141.2 GB532
...ama 3 70B Arimas Story RP V1.6256K / 141.2 GB70
Yi 70B 200K RPMerge Franken195K / 142.4 GB231
Note: green Score (e.g. "73.2") means that the model is better than sophosympatheia/Midnight-Miqu-70B-v1.0.

Rank the Midnight Miqu 70B V1.0 Capabilities

๐Ÿ†˜ 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! ๐ŸŒŸ

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? 45269 in total.

Our Social Media →  
Original data from HuggingFace, OpenCompass and various public git repos.
Release v20241227