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

Breeze 13B 32K Instruct V1.0 by win10

 »  All LLMs  »  win10  »  Breeze 13B 32K Instruct V1.0   URL Share it on

  17   
  Merged Model   Autotrain compatible   Conversational   Endpoints compatible   Instruct Mediatek-research/breeze-7b-32...   Mistral   Region:us   Safetensors   Sharded   Tensorflow
Model Card on HF 🤗: https://huggingface.co/win10/Breeze-13B-32k-Instruct-v1_0 

Breeze 13B 32K Instruct V1.0 Benchmarks

MMLU Pro: 17.42
GPQA: 1.9
MUSR: 11.06
BBH: 25.26
IFEval: 35.84 vs 88 (so35)-59.3%
MATH Lvl 5: 0.98
LLME Score: 0.25025
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
Breeze 13B 32K Instruct V1.0 (win10/Breeze-13B-32k-Instruct-v1_0)

Breeze 13B 32K Instruct V1.0 Parameters and Internals

Model Type 
merge, mergekit, lazymergekit
Additional Notes 
This model was created by merging multiple instances of the MediaTek-Research/Breeze-7B-32k-Instruct-v1_0 model using the MergeKit tool, with passthrough merging technique and specific layer range configuration.
LLM NameBreeze 13B 32K Instruct V1.0
Repository 🤗https://huggingface.co/win10/Breeze-13B-32k-Instruct-v1_0 
Merged ModelYes
Model Size13b
Required VRAM25.6 GB
Updated2024-12-22
Maintainerwin10
Model Typemistral
Instruction-BasedYes
Model Files  0.5 GB: 1-of-27   0.9 GB: 2-of-27   0.9 GB: 3-of-27   1.0 GB: 4-of-27   1.0 GB: 5-of-27   1.0 GB: 6-of-27   0.9 GB: 7-of-27   1.0 GB: 8-of-27   1.0 GB: 9-of-27   1.0 GB: 10-of-27   1.0 GB: 11-of-27   1.0 GB: 12-of-27   1.0 GB: 13-of-27   1.0 GB: 14-of-27   1.0 GB: 15-of-27   1.0 GB: 16-of-27   1.0 GB: 17-of-27   1.0 GB: 18-of-27   0.9 GB: 19-of-27   1.0 GB: 20-of-27   1.0 GB: 21-of-27   1.0 GB: 22-of-27   1.0 GB: 23-of-27   1.0 GB: 24-of-27   1.0 GB: 25-of-27   1.0 GB: 26-of-27   0.5 GB: 27-of-27
Model ArchitectureMistralForCausalLM
Licenseapache-2.0
Context Length32768
Model Max Length32768
Transformers Version4.41.2
Tokenizer ClassLlamaTokenizer
Padding Token</s>
Vocabulary Size62464
Torch Data Typebfloat16

Rank the Breeze 13B 32K Instruct V1.0 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? 40066 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