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

Goku 8x22B V0.2 by MaziyarPanahi

 »  All LLMs  »  MaziyarPanahi  »  Goku 8x22B V0.2   URL Share it on

  28   
  Autotrain compatible   Axolotl Base model:finetune:v2ray/mixt... Base model:v2ray/mixtral-8x22b...   Conversational Dataset:maziyarpanahi/wizardlm... Dataset:microsoft/orca-math-wo...   Dataset:teknium/openhermes-2.5   De   En   Es   Fr   Instruct   It   Mixtral   Moe   Region:us   Safetensors   Sharded   Sharegpt   Tensorflow
Model Card on HF 🤗: https://huggingface.co/MaziyarPanahi/Goku-8x22B-v0.2 

Goku 8x22B V0.2 Benchmarks

LLME Score: 0.19248
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
Goku 8x22B V0.2 (MaziyarPanahi/Goku-8x22B-v0.2)

Goku 8x22B V0.2 Parameters and Internals

Model Type 
text-generation
Additional Notes 
This model has a total of 141b parameters with 35b only active.
Supported Languages 
fr (unknown), it (unknown), de (unknown), es (unknown), en (unknown)
Training Details 
Data Sources:
MaziyarPanahi/WizardLM_evol_instruct_V2_196k, microsoft/orca-math-word-problems-200k, teknium/OpenHermes-2.5
Context Length:
8192
LLM NameGoku 8x22B V0.2
Repository 🤗https://huggingface.co/MaziyarPanahi/Goku-8x22B-v0.2 
Model NameGoku-8x22B-v0.2
Model CreatorMaziyarPanahi
Base Model(s)  v2ray/Mixtral-8x22B-v0.1   v2ray/Mixtral-8x22B-v0.1
Model Size140.6b
Required VRAM211.2 GB
Updated2024-12-21
MaintainerMaziyarPanahi
Model Typemixtral
Instruction-BasedYes
Model Files  4.8 GB: 1-of-117   4.8 GB: 2-of-117   4.8 GB: 3-of-117   4.8 GB: 4-of-117   4.8 GB: 5-of-117   4.8 GB: 6-of-117   4.8 GB: 7-of-117   4.8 GB: 8-of-117   4.8 GB: 9-of-117   4.8 GB: 10-of-117   4.8 GB: 11-of-117   4.8 GB: 12-of-117   4.8 GB: 13-of-117   4.8 GB: 14-of-117   4.8 GB: 15-of-117   4.8 GB: 16-of-117   4.8 GB: 17-of-117   4.8 GB: 18-of-117   4.8 GB: 19-of-117   4.8 GB: 20-of-117   5.0 GB: 21-of-117   4.6 GB: 22-of-117   4.8 GB: 23-of-117   4.8 GB: 24-of-117   4.8 GB: 25-of-117   4.8 GB: 26-of-117   4.8 GB: 27-of-117   4.8 GB: 28-of-117   4.8 GB: 29-of-117   4.8 GB: 30-of-117   4.8 GB: 31-of-117   4.8 GB: 32-of-117   4.8 GB: 33-of-117   4.8 GB: 34-of-117   4.8 GB: 35-of-117   4.8 GB: 36-of-117   4.8 GB: 37-of-117   4.8 GB: 38-of-117   4.8 GB: 39-of-117   4.8 GB: 40-of-117   4.8 GB: 41-of-117   4.8 GB: 42-of-117   4.8 GB: 43-of-117   4.8 GB: 44-of-117
Supported Languagesfr it de es en
Model ArchitectureMixtralForCausalLM
Licenseapache-2.0
Context Length65536
Model Max Length65536
Transformers Version4.40.0.dev0
Vocabulary Size32002
Torch Data Typefloat32

Best Alternatives to Goku 8x22B V0.2

Best Alternatives
Context / RAM
Downloads
Likes
Mixtral 8x22B Instruct V0.164K / 221.4 GB2747604696
...ixtral 8x22B Instruct V0.1 FP864K / 140.9 GB5702
...igHuggyD Grey WizardLM 2 8x22B64K / 216.6 GB194
WizardLM 2 8x22B Beige64K / 221.4 GB243
...x22B Instruct V0.1 FP8 Dynamic64K / 140.9 GB220
...ral 8x22B Instruct V0.1 FP8 V264K / 140.9 GB170
...ral 8x22B Instruct V0.1 FP8 V164K / 140.9 GB100
Mixtral 8x22b Instruct Oh64K / 221.6 GB4329
Mixtral 8x22B GO Instruct V164K / 280.8 GB192
Mixtral 8x22B Instruct V0.164K / 221.4 GB111
Note: green Score (e.g. "73.2") means that the model is better than MaziyarPanahi/Goku-8x22B-v0.2.

Rank the Goku 8x22B V0.2 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