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

Frostwind 10.7B V1 EXL2 by R136a1

 »  All LLMs  »  R136a1  »  Frostwind 10.7B V1 EXL2   URL Share it on

  10   
  Autotrain compatible   En   Endpoints compatible   Exl2   Llama   Quantized   Region:us   Safetensors
Model Card on HF 🤗: https://huggingface.co/R136a1/Frostwind-10.7B-v1-exl2 

Frostwind 10.7B V1 EXL2 Benchmarks

ARC: 64.16 vs 96.7 (so35)-33.7%
HellaSwag: 85.38 vs 95.3 (gpt4)-10.4%
MMLU: 64.64 vs 88.3 (so35)-26.8%
TruthfulQA: 50.43 vs 59 (gpt4)-14.5%
WinoGrande: 83.74 vs 87.5 (gpt4)-4.3%
GSM8K: 52.16 vs 96.4 (so35)-45.9%
LLME Score: 0.16502
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
Frostwind 10.7B V1 EXL2 (R136a1/Frostwind-10.7B-v1-exl2)

Frostwind 10.7B V1 EXL2 Parameters and Internals

Model Type 
Text Generation
Use Cases 
Areas:
Roleplay, Textual Instructive Tasks
Primary Use Cases:
Roleplay scenarios
Considerations:
Needs context to avoid 'as an AI' issues.
Additional Notes 
Fairly smart with filled context. Needs improvisation in roles.
Supported Languages 
en (Fluent)
Training Details 
Data Sources:
Coding, Instruct, Generalised Data, Roleplay, Dealignment
Data Volume:
52K samples
Methodology:
Finetuning with Alpaca format
Training Time:
3 Hours
Hardware Used:
4x 4090 GPUs
LLM NameFrostwind 10.7B V1 EXL2
Repository 🤗https://huggingface.co/R136a1/Frostwind-10.7B-v1-exl2 
Base Model(s)  Sao10K/Frostwind-10.7B-v1   Sao10K/Frostwind-10.7B-v1
Model Size10.7b
Required VRAM10.9 GB
Updated2025-02-22
MaintainerR136a1
Model Typellama
Model Files  10.9 GB
Supported Languagesen
Quantization Typeexl2
Model ArchitectureLlamaForCausalLM
Licensecc-by-nc-4.0
Context Length4096
Model Max Length4096
Transformers Version4.37.0.dev0
Tokenizer ClassLlamaTokenizer
Padding Token</s>
Vocabulary Size32000
Torch Data Typebfloat16

Best Alternatives to Frostwind 10.7B V1 EXL2

Best Alternatives
Context / RAM
Downloads
Likes
... 10.7B Instruct V1.0 128K EXL2128K / 5.6 GB371
...e KoSoLAR 10.7B V0.2 1.4 Dedup4K / 43.2 GB20280
KoSoLAR 10.7B V0.2 1.4 Dedup 14K / 21.6 GB20220
Solar Test 240517 16bit4K / 21.4 GB2430
KoSoLAR 10.7B V0.2 1.4 Dedup4K / 21.6 GB320
Eclectus 1.1 Dedup4K / 21.4 GB20170
Eclectus1.14K / 21.4 GB20140
Solar 10.7B Bnb 4bit4K / 5.9 GB1313
Kiwi Solar Merge Ties2 DPO4K / 21.4 GB1780
...ess 10.7B V1.5B 8.0bpw H8 EXL24K / 10.9 GB41
Note: green Score (e.g. "73.2") means that the model is better than R136a1/Frostwind-10.7B-v1-exl2.

Rank the Frostwind 10.7B V1 EXL2 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? 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