Qwen2 7B Instruct FP8 by neuralmagic

 ยป  All LLMs  ยป  neuralmagic  ยป  Qwen2 7B Instruct FP8   URL Share it on

  Autotrain compatible   Conversational   Endpoints compatible   Fp8   Instruct   Qwen2   Region:us   Safetensors   Sharded   Tensorflow   Vllm

Qwen2 7B Instruct FP8 Benchmarks

nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
Qwen2 7B Instruct FP8 (neuralmagic/Qwen2-7B-Instruct-FP8)

Qwen2 7B Instruct FP8 Parameters and Internals

Model Type 
Text
Use Cases 
Areas:
commercial, research
Primary Use Cases:
assistant-like chat
Limitations:
Use in languages other than English, Use in violation of applicable laws
Additional Notes 
Quantized using AutoFP8 with calibration samples from ultrachat.
Supported Languages 
languages_supported (English), proficiency_level ()
Training Details 
Data Sources:
ultrachat
Data Volume:
512 sequences
Methodology:
AutoFP8 quantization
Context Length:
4096
Model Architecture:
Qwen2
Input Output 
Input Format:
Text
Accepted Modalities:
Text
Output Format:
Text
Performance Tips:
Quantization reduces memory usage by 50%.
Release Notes 
Version:
1.0
Date:
2024-06-14
Notes:
Quantized weights and activations to FP8 for efficiency with vLLM.
LLM NameQwen2 7B Instruct FP8
Repository ๐Ÿค—https://huggingface.co/neuralmagic/Qwen2-7B-Instruct-FP8 
Model Size7b
Required VRAM8.7 GB
Updated2025-01-20
Maintainerneuralmagic
Model Typeqwen2
Instruction-BasedYes
Model Files  5.0 GB: 1-of-2   3.7 GB: 2-of-2
Model ArchitectureQwen2ForCausalLM
Licenseapache-2.0
Context Length32768
Model Max Length32768
Transformers Version4.41.2
Tokenizer ClassQwen2Tokenizer
Padding Token<|endoftext|>
Vocabulary Size152064
Torch Data Typebfloat16
Errorsreplace

Best Alternatives to Qwen2 7B Instruct FP8

Best Alternatives
Context / RAM
Downloads
Likes
Gte Qwen2 7B Instruct128K / 30.5 GB41406245
Rombos LLM V2.5 Qwen 7B128K / 15.2 GB74915
Rombo LLM V2.5 Qwen 7B128K / 15.2 GB311
Cybertron V4 Qw7B UNAMGS128K / 15.2 GB1038
Qwen2 7B Instruct V0.1128K / 15.2 GB409681
Qwen2 7B Instruct V0.8128K / 15.2 GB409903
Model 3128K / 15.2 GB260
Cybertron V4 Qw7B MGS128K / 15.2 GB7811
...eCyberMix Qwen 2.5 7B Instruct128K / 15.2 GB492
UwU 7B Instruct128K / 15.2 GB26128
Note: green Score (e.g. "73.2") means that the model is better than neuralmagic/Qwen2-7B-Instruct-FP8.

Rank the Qwen2 7B Instruct FP8 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? 41636 in total.

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