Frostwind 10.7B V1 AWQ by TheBloke

 ยป  All LLMs  ยป  TheBloke  ยป  Frostwind 10.7B V1 AWQ   URL Share it on

  4-bit   Autotrain compatible   Awq Base model:quantized:sao10k/fr... Base model:sao10k/frostwind-10...   En   Llama   Quantized   Region:us   Safetensors

Frostwind 10.7B V1 AWQ Benchmarks

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 AWQ (TheBloke/Frostwind-10.7B-v1-AWQ)

Frostwind 10.7B V1 AWQ Parameters and Internals

Model Type 
solar
Additional Notes 
Fairly smart, not at the level of bigger models. Needed 1/2 swipes for correct answer in zero context tests. Handles group characters well in roleplays.
Supported Languages 
en (high)
Training Details 
Data Sources:
20% - Coding, 30% - Instruct, 30% - Generalised Data, 10% - Roleplay, 10% - Dealignment
Data Volume:
52K varied samples
Methodology:
fine-tuned from upstage/SOLAR-10.7B-v1.0
Training Time:
Roughly 3 Hours with 4x 4090s
Hardware Used:
4x 4090 GPUs
Model Architecture:
fine-tune
LLM NameFrostwind 10.7B V1 AWQ
Repository ๐Ÿค—https://huggingface.co/TheBloke/Frostwind-10.7B-v1-AWQ 
Model NameFrostwind 10.7B V1
Model CreatorSaofiq
Base Model(s)  Sao10K/Frostwind-10.7B-v1   Sao10K/Frostwind-10.7B-v1
Model Size10.7b
Required VRAM6 GB
Updated2024-12-21
MaintainerTheBloke
Model Typesolar
Model Files  6.0 GB
Supported Languagesen
AWQ QuantizationYes
Quantization Typeawq
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 Typefloat16

Best Alternatives to Frostwind 10.7B V1 AWQ

Best Alternatives
Context / RAM
Downloads
Likes
STOCK SOLAR 10.7B AWQ4K / 6 GB140
KULLM3 AWQ4K / 6 GB1205
Tess 10.7B V1.5B AWQ4K / 6 GB284
Nous Hermes 2 SOLAR 10.7B AWQ4K / 6 GB3910
SOLAR 10.7B Instruct V1.0 AWQ4K / 6 GB1995
Sensualize Solar 10.7B AWQ4K / 6 GB242
CarbonVillain En 10.7B V4 AWQ4K / 6 GB121
LMCocktail 10.7B V1 AWQ4K / 6 GB191
Solar 10.7B SLERP AWQ4K / 6 GB411
SOLAR Platypus 10.7B V2 AWQ4K / 6 GB222
Note: green Score (e.g. "73.2") means that the model is better than TheBloke/Frostwind-10.7B-v1-AWQ.

Rank the Frostwind 10.7B V1 AWQ 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? 40013 in total.

Our Social Media →  
Original data from HuggingFace, OpenCompass and various public git repos.
Release v20241217