Mantis 8B Fuyu by TIGER-Lab

 ยป  All LLMs  ยป  TIGER-Lab  ยป  Mantis 8B Fuyu   URL Share it on

  Arxiv:2405.01483   Autotrain compatible   Base model:adept/fuyu-8b Base model:finetune:adept/fuyu... Dataset:tiger-lab/mantis-instr...   En   Endpoints compatible   Fuyu   Instruct   Lmm   Mantis   Mfuyu   Multimodal   Region:us   Safetensors   Sharded   Tensorflow   Vlm

Mantis 8B Fuyu Benchmarks

nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
Mantis 8B Fuyu (TIGER-Lab/Mantis-8B-Fuyu)

Mantis 8B Fuyu Parameters and Internals

Model Type 
multimodal, lmm, vlm
Use Cases 
Areas:
research, academic, non-commercial
Primary Use Cases:
multi-image tasks, image reasoning, temporal understanding
Limitations:
Not suitable for commercial use
Additional Notes 
Achieves state-of-the-art multi-image skills including co-reference, reasoning, comparing, temporal understanding.
Supported Languages 
en (high)
Training Details 
Data Sources:
TIGER-Lab/Mantis-Instruct
Training Time:
36 hours
Hardware Used:
16xA100-40G
Model Architecture:
Interleaved text and image as inputs with multi-image capabilities
Input Output 
Input Format:
Interleaved text and images
LLM NameMantis 8B Fuyu
Repository ๐Ÿค—https://huggingface.co/TIGER-Lab/Mantis-8B-Fuyu 
Base Model(s)  Fuyu 8B   adept/fuyu-8b
Model Size8b
Required VRAM18.9 GB
Updated2025-02-22
MaintainerTIGER-Lab
Model Typefuyu
Instruction-BasedYes
Model Files  5.0 GB: 1-of-4   5.0 GB: 2-of-4   5.0 GB: 3-of-4   3.9 GB: 4-of-4   0.0 GB
Supported Languagesen
Model ArchitectureMFuyuForCausalLM
Licensecc-by-nc-4.0
Context Length16384
Model Max Length16384
Transformers Version4.39.2
Tokenizer ClassLlamaTokenizer
Vocabulary Size262146
Torch Data Typebfloat16

Rank the Mantis 8B Fuyu Capabilities

๐Ÿ†˜ 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! ๐ŸŒŸ

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? 43470 in total.

Our Social Media →  
Original data from HuggingFace, OpenCompass and various public git repos.
Release v20241227