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

ConvLLaVA JP 1.3B 768 by toshi456

 »  All LLMs  »  toshi456  »  ConvLLaVA JP 1.3B 768   URL Share it on

  18   
  Arxiv:2405.15738   Autotrain compatible Dataset:turing-motors/llava-pr... Dataset:turing-motors/llava-v1...   Endpoints compatible   Image-captioning   Image-to-text   Instruct   Ja   Llava-jp   Region:us   Safetensors   Sharded   Tensorflow   Vision   Vqa
Model Card on HF 🤗: https://huggingface.co/toshi456/ConvLLaVA-JP-1.3b-768 

ConvLLaVA JP 1.3B 768 Benchmarks

LLME Score: 0.19119
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
ConvLLaVA JP 1.3B 768 (toshi456/ConvLLaVA-JP-1.3b-768)

ConvLLaVA JP 1.3B 768 Parameters and Internals

Model Type 
vision-language, image-to-text, VQA, image-captioning
Additional Notes 
ConvLLaVA-JP is trained using the laion/CLIP-convnext_large_d_320.laion2B-s29B-b131K-ft image encoder and llm-jp/llm-jp-1.3b-v1.0 text decoder. Input resolution: 768 x 768.
Supported Languages 
ja (Native)
Training Details 
Data Sources:
LLaVA-Pretrain-JA, LLaVA-v1.5-Instruct-620K-JA
Input Output 
Accepted Modalities:
image
Output Format:
text
LLM NameConvLLaVA JP 1.3B 768
Repository 🤗https://huggingface.co/toshi456/ConvLLaVA-JP-1.3b-768 
Model Size1.3b
Required VRAM7.1 GB
Updated2025-02-22
Maintainertoshi456
Model Typellava-jp
Instruction-BasedYes
Model Files  5.0 GB: 1-of-2   2.1 GB: 2-of-2
Supported Languagesja
Model ArchitectureLlavaGpt2ForCausalLM
Licensecc-by-nc-4.0
Model Max Length1532
Transformers Version4.38.2
Tokenizer ClassPreTrainedTokenizerFast
Padding Token<unk|LLM-jp>
Vocabulary Size50688
Torch Data Typefloat32
Activation Functiongelu

Best Alternatives to ConvLLaVA JP 1.3B 768

Best Alternatives
Context / RAM
Downloads
Likes
Llava Jp 1.3B V1.10K / 6.6 GB70511
ConvLLaVA JP 1.3B 12800K / 7.1 GB191
...3B V1.1 Llava Jp Instruct 108K0K / 6.6 GB93
...V1.0 Siglip So400m Patch14 3840K / 6.6 GB650
Llava Jp 1.3B V1.00K / 6.3 GB735
Note: green Score (e.g. "73.2") means that the model is better than toshi456/ConvLLaVA-JP-1.3b-768.

Rank the ConvLLaVA JP 1.3B 768 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? 43470 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