Qwen2 VL 2B Instruct GPTQ Int8 by Qwen

 ยป  All LLMs  ยป  Qwen  ยป  Qwen2 VL 2B Instruct GPTQ Int8   URL Share it on

  Arxiv:2308.12966   Arxiv:2409.12191   8-bit Base model:quantized:qwen/qwen... Base model:qwen/qwen2-vl-2b-in...   Conversational   En   Gptq   Image-text-to-text   Instruct   Multimodal   Quantized   Qwen2 vl   Region:us   Safetensors

Qwen2 VL 2B Instruct GPTQ Int8 Benchmarks

nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
Qwen2 VL 2B Instruct GPTQ Int8 (Qwen/Qwen2-VL-2B-Instruct-GPTQ-Int8)

Qwen2 VL 2B Instruct GPTQ Int8 Parameters and Internals

Model Type 
text generation, multimodal
Use Cases 
Areas:
commercial applications, research
Applications:
visual question answering, dialog systems, content creation, device integration
Primary Use Cases:
Visual QA, Dialog, Content Creation
Limitations:
Lack of Audio Support, Data timeliness until June 2023, Recognition of specific individuals or IPs, Limited handling of complex instructions, Insufficient counting accuracy, Weak spatial reasoning skills
Additional Notes 
Available in different quantizations for broad hardware compatibility.
Supported Languages 
en (>=0.8), zh (>=0.8), fr (>=0.8), es (>=0.8), de (>=0.8), ru (>=0.8), ja (>=0.8), ko (>=0.8), ar (>=0.8), vi (>=0.8)
Training Details 
Data Sources:
MathVista, DocVQA, RealWorldQA, MTVQA
Methodology:
Instruction-tuning, GPTQ quantization
Model Architecture:
Naive Dynamic Resolution, Multimodal Rotary Position Embedding (M-ROPE)
Input Output 
Input Format:
Message-based input with role specification
Accepted Modalities:
text, image, video
Output Format:
Textual descriptions or responses
Performance Tips:
Use flash_attention_2 for acceleration in multi-image and video scenarios
Release Notes 
Version:
2B
Notes:
Quantized relay in 2B format, instruction-tuned.
LLM NameQwen2 VL 2B Instruct GPTQ Int8
Repository ๐Ÿค—https://huggingface.co/Qwen/Qwen2-VL-2B-Instruct-GPTQ-Int8 
Base Model(s)  Qwen2 VL 2B Instruct   Qwen/Qwen2-VL-2B-Instruct
Model Size2b
Required VRAM3.1 GB
Updated2025-04-22
MaintainerQwen
Model Typeqwen2_vl
Instruction-BasedYes
Model Files  3.1 GB
Supported Languagesen
GPTQ QuantizationYes
Quantization Typegptq
Model ArchitectureQwen2VLForConditionalGeneration
Licenseapache-2.0
Context Length32768
Model Max Length32768
Transformers Version4.45.0.dev0
Tokenizer ClassQwen2Tokenizer
Padding Token<|endoftext|>
Vocabulary Size151936
Torch Data Typefloat16
Errorsreplace

Best Alternatives to Qwen2 VL 2B Instruct GPTQ Int8

Best Alternatives
Context / RAM
Downloads
Likes
Qwen2 VL 2B Instruct GPTQ Int432K / 2.5 GB660424
Qwen2 VL 2B Instruct AWQ32K / 3 GB239822
Qwen2 VL 2B Instruct32K / 4.4 GB324
Note: green Score (e.g. "73.2") means that the model is better than Qwen/Qwen2-VL-2B-Instruct-GPTQ-Int8.

Rank the Qwen2 VL 2B Instruct GPTQ Int8 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? 46565 in total.

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