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

ChatWaifu 22B V2.0 Preview by spow12

 »  All LLMs  »  spow12  »  ChatWaifu 22B V2.0 Preview   URL Share it on

  29   
  Autotrain compatible Base model:finetune:mistralai/... Base model:mistralai/mistral-s...   Conversational Dataset:aratako/synthetic-japa... Dataset:aratako/synthetic-japa... Dataset:gryphe/sonnet3.5-slimo... Dataset:kalomaze/opus instruct... Dataset:roleplay4fun/aesir-v1.... Dataset:skunkworksai/reasoning...   En   Endpoints compatible   Instruct   Ja   Merge   Mergekit   Mistral   Model-index   Nsfw   Region:us   Roleplay   Safetensors   Sharded   Tensorflow   Visual novel
Model Card on HF 🤗: https://huggingface.co/spow12/ChatWaifu_22B_v2.0_preview 

ChatWaifu 22B V2.0 Preview Benchmarks

MMLU Pro: 33.2
GPQA: 8.72
MUSR: 3.53
BBH: 45.49
IFEval: 67.45 vs 88 (so35)-23.4%
MATH Lvl 5: 18.05
LLME Score: 0.32007
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
ChatWaifu 22B V2.0 Preview (spow12/ChatWaifu_22B_v2.0_preview)

ChatWaifu 22B V2.0 Preview Parameters and Internals

Model Type 
CausalLM, text-generation
Use Cases 
Areas:
Research, Text-generation
Applications:
Visual novel roleplay, Character-based interactions
Primary Use Cases:
NSFW Content Generation, Visual Novel Character Simulation
Limitations:
Can generate NSFW content.
Considerations:
Use in settings suitable for NSFW content, avoid problematic dialogues or realistic personal interactions.
Additional Notes 
Merges models incorporating anime characters for text-based interaction with specific user-generated character personas in mind.
Supported Languages 
languages_supported (Japanese, English), proficiency (Fluent/)
Training Details 
Data Sources:
roleplay4fun/aesir-v1.1, kalomaze/Opus_Instruct_3k, Gryphe/Sonnet3.5-SlimOrcaDedupCleaned, Aratako/Synthetic-Japanese-Roleplay-gpt-4o-mini-39.6k-formatted, Aratako/Synthetic-Japanese-Roleplay-NSFW-Claude-3.5s-15.3k-formatted, SkunkworksAI/reasoning-0.01
Data Volume:
Approximately 50,000 samples from Aratako/Synthetic-JP-EN-Coding-Dataset-567k.
Methodology:
Merge of models using slerp method.
Context Length:
128000
Hardware Used:
bfloat16
Input Output 
Input Format:
Text including persona descriptions.
Accepted Modalities:
text
Output Format:
Text character simulations
Performance Tips:
Ensure clean context management for optimal character consistency.
Release Notes 
Version:
2.0
Date:
2024-09-23
Notes:
Updated to 22B version with Ver 2.0 improvements.
LLM NameChatWaifu 22B V2.0 Preview
Repository 🤗https://huggingface.co/spow12/ChatWaifu_22B_v2.0_preview 
Base Model(s)  Mistral Small Instruct 2409   mistralai/Mistral-Small-Instruct-2409
Model Size22b
Required VRAM44.7 GB
Updated2024-12-21
Maintainerspow12
Model Typemistral
Instruction-BasedYes
Model Files  4.9 GB: 1-of-9   5.0 GB: 2-of-9   5.0 GB: 3-of-9   4.9 GB: 4-of-9   5.0 GB: 5-of-9   5.0 GB: 6-of-9   4.9 GB: 7-of-9   5.0 GB: 8-of-9   5.0 GB: 9-of-9
Supported Languagesen ja
Model ArchitectureMistralForCausalLM
Licensecc-by-nc-4.0
Context Length32768
Model Max Length32768
Transformers Version4.44.2
Tokenizer ClassLlamaTokenizer
Padding Token<unk>
Vocabulary Size32768
Torch Data Typebfloat16

Best Alternatives to ChatWaifu 22B V2.0 Preview

Best Alternatives
Context / RAM
Downloads
Likes
MS Schisandra 22B V0.2128K / 44.7 GB797
...ntheon RP Pure 1.6.2 22B Small128K / 44.7 GB8117
MS Meadowlark 22B128K / 44.7 GB1119
Cydonia 22B V1128K / 44.7 GB11052
MSM MS Cydrion 22B128K / 44.7 GB9616
Pantheon RP 1.6.2 22B Small128K / 44.7 GB5211
MS Meadowlark Alt 22B128K / 44.7 GB301
UnslopSmall 22B V1128K / 44.7 GB305
Acolyte 22B128K / 44.7 GB238
SeminalRP 22B128K / 44.7 GB392
Note: green Score (e.g. "73.2") means that the model is better than spow12/ChatWaifu_22B_v2.0_preview.

Rank the ChatWaifu 22B V2.0 Preview 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? 40013 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.
Release v20241217
  Support LLM Explorer