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

Kunoichi Lemon Royale V2 32K 7B by grimjim

 »  All LLMs  »  grimjim  »  Kunoichi Lemon Royale V2 32K 7B   URL Share it on

  6   

Kunoichi Lemon Royale V2 32K 7B is an open-source language model by grimjim. Features: 7b LLM, VRAM: 14.5GB, Context: 32K, License: cc-by-nc-4.0, Instruction-Based, Merged, HF Score: 67.5, LLM Explorer Score: 0.14, Arc: 66.6, HellaSwag: 85.3, MMLU: 62.9, TruthfulQA: 58.2, WinoGrande: 78.7, GSM8K: 53.5.

  Merged Model   Arxiv:2212.04089 Base model:grimjim/kunoichi-le... Base model:grimjim/mistral-7b-...   Conversational   Endpoints compatible   Instruct   Mistral   Region:us   Safetensors   Sharded   Tensorflow
Model Card on HF 🤗: https://huggingface.co/grimjim/kunoichi-lemon-royale-v2-32K-7B 

Kunoichi Lemon Royale V2 32K 7B Benchmarks

ARC: 66.55 vs 96.7 (so35)-31.2%
HellaSwag: 85.25 vs 95.3 (gpt4)-10.5%
MMLU: 62.93 vs 88.3 (so35)-28.7%
TruthfulQA: 58.23 vs 59 (gpt4)-1.3%
WinoGrande: 78.69 vs 87.5 (gpt4)-10.1%
GSM8K: 53.45 vs 96.4 (so35)-44.6%
LLME Score: 0.13844
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
Kunoichi Lemon Royale V2 32K 7B (grimjim/kunoichi-lemon-royale-v2-32K-7B)
🌟 Advertise your project 🚀

Kunoichi Lemon Royale V2 32K 7B Parameters and Internals

Model Type 
text-generation
Additional Notes 
This model was merged from models derived from Mistral v0.1 and v0.2 using the task arithmetic method. Rope_theta in config.json was adjusted to work well to 16K context length.
Training Details 
Methodology:
task_arithmetic
Context Length:
32000
LLM NameKunoichi Lemon Royale V2 32K 7B
Repository 🤗https://huggingface.co/grimjim/kunoichi-lemon-royale-v2-32K-7B 
Base Model(s)  grimjim/Mistral-7B-Instruct-demi-merge-v0.2-7B   Kunoichi Lemon Royale 7B   grimjim/Mistral-7B-Instruct-demi-merge-v0.2-7B   grimjim/kunoichi-lemon-royale-7B
Merged ModelYes
Model Size7b
Required VRAM14.5 GB
Updated2026-04-30
Maintainergrimjim
Model Typemistral
Instruction-BasedYes
Model Files  5.9 GB: 1-of-3   6.0 GB: 2-of-3   2.6 GB: 3-of-3
Model ArchitectureMistralForCausalLM
Licensecc-by-nc-4.0
Context Length32768
Model Max Length32768
Transformers Version4.39.2
Tokenizer ClassLlamaTokenizer
Vocabulary Size32000
Torch Data Typebfloat16

Best Alternatives to Kunoichi Lemon Royale V2 32K 7B

Best Alternatives
Context / RAM
Downloads
Likes
...Nemo Instruct 2407 Abliterated1000K / 24.5 GB25420
SpydazWeb AI HumanAI RP512K / 14.4 GB211
SpydazWeb AI HumanAI 002512K / 14.4 GB181
...daz Web AI ChatML 512K Project512K / 14.5 GB120
... Summarize 64K QLoRANET Merged128K / 4.1 GB60
...1 Summarize 64K LoRANET Merged128K / 14.4 GB60
Mistral 7B Instruct V0.232K / 14.4 GB20675793127
Mistral 7B Instruct V0.132K / 14.4 GB4266851832
Mistral 7B Instruct V0.232K / 14.4 GB57590
Ghost V5 Ultra 8B32K / 14.5 GB990
Note: green Score (e.g. "73.2") means that the model is better than grimjim/kunoichi-lemon-royale-v2-32K-7B.

Rank the Kunoichi Lemon Royale V2 32K 7B 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? 53373 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