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

Starling LM 7B Alpha by berkeley-nest

 »  All LLMs  »  berkeley-nest  »  Starling LM 7B Alpha   URL Share it on

  17136   
  Arxiv:2306.02231   Autotrain compatible   Conversational   Dataset:berkeley-nest/nectar   En   Endpoints compatible   Mistral   Region:us   Reward model   Rlaif   Rlhf   Safetensors   Sharded   Tensorflow
Model Card on HF 🤗: https://huggingface.co/berkeley-nest/Starling-LM-7B-alpha 

Starling LM 7B Alpha Benchmarks

LMSys ELO: 1088 vs 1272 (so35)-14.5%
MMLU Pro: 24.13
GPQA: 6.26
MUSR: 9.5
BBH: 21.95
IFEval: 54.8 vs 88 (so35)-37.7%
ARC: 63.65 vs 96.7 (so35)-34.2%
HellaSwag: 84.87 vs 95.3 (gpt4)-10.9%
MMLU: 64.7 vs 88.3 (so35)-26.7%
TruthfulQA: 46.32 vs 59 (gpt4)-21.5%
WinoGrande: 80.43 vs 87.5 (gpt4)-8.1%
GSM8K: 62.32 vs 96.4 (so35)-35.4%
MATH Lvl 5: 8.38
LLME Score: 0.38132
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
Starling LM 7B Alpha (berkeley-nest/Starling-LM-7B-alpha)

Starling LM 7B Alpha Parameters and Internals

Model Type 
Language Model, Reward Model
Use Cases 
Areas:
Research, Non-commercial use
Considerations:
Stay tuned for our forthcoming code and paper for details
Additional Notes 
An online demo is available in LMSYS Chatbot Arena
Supported Languages 
en (Proficient)
Training Details 
Data Sources:
berkeley-nest/Nectar
Methodology:
Reinforcement Learning from AI Feedback (RLAIF)
Input Output 
Input Format:
Follows the exact chat template and usage as Openchat 3.5.
Accepted Modalities:
Text
Output Format:
Generated text
Performance Tips:
Ensure to use the exact chat template for optimal performance; set temperature = 0 to reduce verbosity.
LLM NameStarling LM 7B Alpha
Repository 🤗https://huggingface.co/berkeley-nest/Starling-LM-7B-alpha 
Model Size7b
Required VRAM14.4 GB
Updated2025-02-22
Maintainerberkeley-nest
Model Typemistral
Model Files  4.9 GB: 1-of-3   5.0 GB: 2-of-3   4.5 GB: 3-of-3
Supported Languagesen
Model ArchitectureMistralForCausalLM
Licenseapache-2.0
Context Length8192
Model Max Length8192
Transformers Version4.35.0
Tokenizer ClassLlamaTokenizer
Padding Token<|end_of_turn|>
Vocabulary Size32002
Torch Data Typebfloat16

Quantized Models of the Starling LM 7B Alpha

Model
Likes
Downloads
VRAM
Starling LM 7B Alpha GGUF9413643 GB
Starling LM 7B Alpha GGUF11492 GB
Starling LM 7B Alpha AWQ9874 GB
Starling LM 7B Alpha GPTQ9384 GB
Starling LM 7B Alpha GGUF3562 GB

Best Alternatives to Starling LM 7B Alpha

Best Alternatives
Context / RAM
Downloads
Likes
...Nemo Instruct 2407 Abliterated1000K / 24.5 GB462011
MegaBeam Mistral 7B 512K512K / 14.4 GB568150
SpydazWeb AI HumanAI RP512K / 14.4 GB121
SpydazWeb AI HumanAI 002512K / 14.4 GB181
...daz Web AI ChatML 512K Project512K / 14.5 GB120
MegaBeam Mistral 7B 300K282K / 14.4 GB563316
Hebrew Mistral 7B 200K256K / 30 GB1461915
Astral 256K 7B V2250K / 14.4 GB70
Astral 256K 7B250K / 14.4 GB50
Test001128K / 14.5 GB90

Rank the Starling LM 7B Alpha 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? 43508 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