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

SOLAR 0 70B 16bit by upstage

 »  All LLMs  »  upstage  »  SOLAR 0 70B 16bit   URL Share it on

  2698   
  16bit   Autotrain compatible   En   Endpoints compatible   Instruct   Instruction   Llama   Llama2   Pytorch   Quantized   Region:us   Sharded   Upstage
Model Card on HF 🤗: https://huggingface.co/upstage/SOLAR-0-70b-16bit 

SOLAR 0 70B 16bit Benchmarks

ARC: 71.08 vs 96.7 (so35)-26.5%
HellaSwag: 87.89 vs 95.3 (gpt4)-7.8%
MMLU: 70.58 vs 88.3 (so35)-20.1%
TruthfulQA: 62.25 vs 59 (gpt4)5.5%
WinoGrande: 83.58 vs 87.5 (gpt4)-4.5%
GSM8K: 45.26 vs 96.4 (so35)-53%
LLME Score: 0.1611
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
SOLAR 0 70B 16bit (upstage/SOLAR-0-70b-16bit)

SOLAR 0 70B 16bit Parameters and Internals

Model Type 
text generation
Use Cases 
Areas:
research, private LLMS, AI applications
Additional Notes 
The model utilizes Dynamic rope_scaling to handle larger input prompts.
Supported Languages 
english (fluent)
Training Details 
Data Sources:
Orca-style dataset, Alpaca-style dataset
Methodology:
Fine-tuning
Context Length:
10000
Hardware Used:
A100x8 * 4
Input Output 
Input Format:
Text-based prompts with specific structure using user and assistant delineations
Accepted Modalities:
text
Output Format:
Text output aligned with input prompts
Performance Tips:
The model can handle over 10,000 input tokens using 'rope_scaling'.
LLM NameSOLAR 0 70B 16bit
Repository 🤗https://huggingface.co/upstage/SOLAR-0-70b-16bit 
Model Size70b
Required VRAM138 GB
Updated2025-05-12
Maintainerupstage
Model Typellama
Instruction-BasedYes
Model Files  9.8 GB: 1-of-15   9.8 GB: 2-of-15   10.0 GB: 3-of-15   9.8 GB: 4-of-15   9.8 GB: 5-of-15   9.8 GB: 6-of-15   10.0 GB: 7-of-15   9.8 GB: 8-of-15   9.8 GB: 9-of-15   9.8 GB: 10-of-15   10.0 GB: 11-of-15   9.8 GB: 12-of-15   9.8 GB: 13-of-15   9.5 GB: 14-of-15   0.5 GB: 15-of-15
Supported Languagesen
Quantization Type16bit
Model ArchitectureLlamaForCausalLM
Context Length4096
Model Max Length4096
Transformers Version4.31.0
Tokenizer ClassLlamaTokenizer
Beginning of Sentence Token<s>
End of Sentence Token</s>
Unk Token<unk>
Vocabulary Size32000
Torch Data Typefloat16

Best Alternatives to SOLAR 0 70B 16bit

Best Alternatives
Context / RAM
Downloads
Likes
...B Instruct Gradient 1048K 8bit1024K / 75 GB125
...B Instruct Gradient 1048K 4bit1024K / 39.7 GB53
...B Instruct Gradient 1048K 2bit1024K / 21.9 GB142
...0B Instruct Gradient 262K 8bit256K / 75 GB142
...0B Instruct Gradient 262K 4bit256K / 39.7 GB53
...0B Instruct Gradient 262K 2bit256K / 21.9 GB51
... Gradient 262K 2.25bpw H6 EXL2256K / 22.2 GB80
...t Gradient 262K 2.4bpw H6 EXL2256K / 23.5 GB130
...t Gradient 262K 4.0bpw H6 EXL2256K / 37.2 GB51
...t Gradient 262K 3.5bpw H6 EXL2256K / 32.9 GB50
Note: green Score (e.g. "73.2") means that the model is better than upstage/SOLAR-0-70b-16bit.

Rank the SOLAR 0 70B 16bit 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? 47272 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