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

Goliath120B EXL2 2 2.64bpw by LavaPlanet

 »  All LLMs  »  LavaPlanet  »  Goliath120B EXL2 2 2.64bpw   URL Share it on

  16   
  Autotrain compatible   Endpoints compatible   Exl2   Llama   Quantized   Region:us   Safetensors   Sharded   Tensorflow
Model Card on HF 🤗: https://huggingface.co/LavaPlanet/Goliath120B-exl2_2-2.64bpw 

Goliath120B EXL2 2 2.64bpw Benchmarks

LLME Score: 0.16123
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
Goliath120B EXL2 2 2.64bpw (LavaPlanet/Goliath120B-exl2_2-2.64bpw)

Goliath120B EXL2 2 2.64bpw Parameters and Internals

Model Type 
text generation
Additional Notes 
EXL2 version of AlpinDale's model with new experimental quant method.
Training Details 
Data Sources:
Pippa llama2 Chat
Methodology:
experimental quant method of exllamav2
Context Length:
4096
Hardware Used:
RTX 3090 (2 units, 24GB VRAM each)
Input Output 
Input Format:
2.64BPW
Accepted Modalities:
text
Output Format:
text
Performance Tips:
Run split across two RTX 3090s; approximately 10 tokens per second throughput with GPU split: 18/24 GPU1: 19.8/24, GPU2: 21.9/24
LLM NameGoliath120B EXL2 2 2.64bpw
Repository 🤗https://huggingface.co/LavaPlanet/Goliath120B-exl2_2-2.64bpw 
Model Size120b
Required VRAM39.5 GB
Updated2024-12-14
MaintainerLavaPlanet
Model Typellama
Model Files  8.6 GB: 1-of-5   8.6 GB: 2-of-5   8.6 GB: 3-of-5   8.5 GB: 4-of-5   5.2 GB: 5-of-5
Quantization Typeexl2
Model ArchitectureLlamaForCausalLM
Licensellama2
Context Length4096
Model Max Length4096
Transformers Version4.35.0
Tokenizer ClassLlamaTokenizer
Vocabulary Size32000
Torch Data Typefloat16

Best Alternatives to Goliath120B EXL2 2 2.64bpw

Best Alternatives
Context / RAM
Downloads
Likes
Miquella 120B 3.0bpw H6 EXL231K / 44.8 GB1410
Miquella 120B 8.0bpw H8 EXL231K / 118.1 GB113
Miquella 120B 4.0bpw H6 EXL231K / 59.4 GB122
...t 120B Cat A Llama EXL2 5.5bpw8K / 85.3 GB150
...t 120B Cat A Llama EXL2 4.5bpw8K / 70.3 GB71
...h LongLORA 120B Rope8 32K Fp164K / 235.4 GB147
...RA 120B Rope8 32K 6bpw H8 EXL24K / 88.7 GB151
...egaDolphin 120B 2.9bpw H6 EXL24K / 44.3 GB153
...gaDolphin 120B 2.65bpw H6 EXL24K / 40.5 GB112
...egaDolphin 120B 4.0bpw H6 EXL24K / 60.8 GB131
Note: green Score (e.g. "73.2") means that the model is better than LavaPlanet/Goliath120B-exl2_2-2.64bpw.

Rank the Goliath120B EXL2 2 2.64bpw 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? 39237 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 v20241124
  Support LLM Explorer