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

MetaMath Cybertron Starling by Q-bert

 »  All LLMs  »  Q-bert  »  MetaMath Cybertron Starling   URL Share it on

  108   

MetaMath Cybertron Starling is an open-source language model by Q-bert. Features: 7b LLM, VRAM: 14.4GB, Context: 32K, License: cc-by-nc-4.0, LLM Explorer Score: 0.12, Arc: 67.4, HellaSwag: 86.3, MMLU: 65.1, GSM8K: 71.5.

Base model:berkeley-nest/starl... Base model:merge:berkeley-nest... Base model:merge:q-bert/metama... Base model:q-bert/metamath-cyb...   Dataset:meta-math/metamathqa   En   Endpoints compatible   Math   Merge   Mistral   Region:us   Safetensors   Sharded   Tensorflow
Model Card on HF 🤗: https://huggingface.co/Q-bert/MetaMath-Cybertron-Starling 

MetaMath Cybertron Starling Benchmarks

ARC: 67.41 vs 96.7 (so35)-30.3%
HellaSwag: 86.26 vs 95.3 (gpt4)-9.5%
MMLU: 65.09 vs 88.3 (so35)-26.3%
GSM8K: 71.49 vs 96.4 (so35)-25.8%
LLME Score: 0.11731
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
MetaMath Cybertron Starling (Q-bert/MetaMath-Cybertron-Starling)
🌟 Advertise your project 🚀

MetaMath Cybertron Starling Parameters and Internals

Model Type 
text-generation
Supported Languages 
en (high)
Input Output 
Input Format:
ChatML format
LLM NameMetaMath Cybertron Starling
Repository 🤗https://huggingface.co/Q-bert/MetaMath-Cybertron-Starling 
Base Model(s)  MetaMath Cybertron   Starling LM 7B Alpha   Q-bert/MetaMath-Cybertron   berkeley-nest/Starling-LM-7B-alpha
Model Size7b
Required VRAM14.4 GB
Updated2026-04-27
MaintainerQ-bert
Model Typemistral
Model Files  9.9 GB: 1-of-2   4.5 GB: 2-of-2
Supported Languagesen
Model ArchitectureMistralForCausalLM
Licensecc-by-nc-4.0
Context Length32768
Model Max Length32768
Transformers Version4.35.2
Tokenizer ClassLlamaTokenizer
Vocabulary Size32000
Torch Data Typebfloat16

Quantized Models of the MetaMath Cybertron Starling

Model
Likes
Downloads
VRAM
...taMath Cybertron Starling GGUF01752 GB
...taMath Cybertron Starling GGUF124683 GB
...taMath Cybertron Starling GPTQ174 GB
...etaMath Cybertron Starling AWQ174 GB

Best Alternatives to MetaMath Cybertron Starling

Best Alternatives
Context / RAM
Downloads
Likes
...Nemo Instruct 2407 Abliterated1000K / 24.5 GB25420
MegaBeam Mistral 7B 512K512K / 14.4 GB862354
SpydazWeb AI HumanAI RP512K / 14.4 GB241
SpydazWeb AI HumanAI 002512K / 14.4 GB181
...daz Web AI ChatML 512K Project512K / 14.5 GB120
MegaBeam Mistral 7B 300K282K / 14.4 GB377916
MegaBeam Mistral 7B 300K282K / 14.4 GB813117
Hebrew Mistral 7B 200K256K / 30 GB137815
Astral 256K 7B V2250K / 14.4 GB50
Astral 256K 7B250K / 14.4 GB50
Note: green Score (e.g. "73.2") means that the model is better than Q-bert/MetaMath-Cybertron-Starling.

Rank the MetaMath Cybertron Starling 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? 53999 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.
Check out Ag3ntum — our secure, self-hosted AI agent for server management.
Release v20260328a
  Support LLM Explorer