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

Ninja V1 by Local-Novel-LLM-project

 Β»  All LLMs  Β»  Local-Novel-LLM-project  Β»  Ninja V1   URL Share it on

  29   
  Autotrain compatible   En   Endpoints compatible   Finetuned   Ja   Mistral   Region:us   Safetensors   Sharded   Tensorflow
Model Card on HF πŸ€—: https://huggingface.co/Local-Novel-LLM-project/Ninja-v1 

Ninja V1 Benchmarks

LLME Score: 0.18278
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
Ninja V1 (Local-Novel-LLM-project/Ninja-v1)

Ninja V1 Parameters and Internals

Model Type 
text generation
Use Cases 
Areas:
research, commercial applications
Primary Use Cases:
High quality Japanese and English generation, Memory ability that does not forget even after long-context generation
Limitations:
The training data may be biased
Considerations:
Be careful with the generated sentences. Memory usage may be large for long inferences.
Additional Notes 
This model includes features from WizardLM2, Antler-7B, and NTQAI/chatntq-ja-7b-v1.0 to enhance multitasking, novel writing, and Japanese quality.
Supported Languages 
en (high quality), ja (high quality)
Training Details 
Data Sources:
noveldataset
Methodology:
Simple linear merging of result models, Domain and Sentence Enhancement with LORA, Context expansion
Context Length:
128000
Hardware Used:
GPUs from the first LocalAI hackathon
Model Architecture:
based on Mistral-7B
Input Output 
Input Format:
Prompt format from Vicuna for multi-turn conversation.
Accepted Modalities:
text
Output Format:
Text
Performance Tips:
Example prompts to improve Japanese generation: BAD: あγͺγŸγ―β—‹β—‹γ¨γ—γ¦ζŒ―γ‚‹θˆžγ„γΎγ™, GOOD: あγͺγŸγ―β—‹β—‹γ§γ™
LLM NameNinja V1
Repository πŸ€—https://huggingface.co/Local-Novel-LLM-project/Ninja-v1 
Model Size7b
Required VRAM14.4 GB
Updated2025-02-22
MaintainerLocal-Novel-LLM-project
Model Typemistral
Model Files  4.9 GB: 1-of-3   5.0 GB: 2-of-3   4.5 GB: 3-of-3
Supported Languagesen ja
Model ArchitectureMistralForCausalLM
Licenseapache-2.0
Context Length4096
Model Max Length4096
Transformers Version4.39.3
Tokenizer ClassLlamaTokenizer
Vocabulary Size32000
Torch Data Typebfloat16

Best Alternatives to Ninja V1

Best Alternatives
Context / RAM
Downloads
Likes
...Nemo Instruct 2407 Abliterated1000K / 24.5 GB462011
MegaBeam Mistral 7B 512K512K / 14.4 GB568150
SpydazWeb AI HumanAI RP512K / 14.4 GB121
SpydazWeb AI HumanAI 002512K / 14.4 GB181
...daz Web AI ChatML 512K Project512K / 14.5 GB120
MegaBeam Mistral 7B 300K282K / 14.4 GB563316
Hebrew Mistral 7B 200K256K / 30 GB1461915
Astral 256K 7B V2250K / 14.4 GB70
Astral 256K 7B250K / 14.4 GB50
Test001128K / 14.5 GB90
Note: green Score (e.g. "73.2") means that the model is better than Local-Novel-LLM-project/Ninja-v1.

Rank the Ninja V1 Capabilities

πŸ†˜ 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? 43470 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 v20241227
  Support LLM Explorer