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

Qwen1.5 14B by Qwen

 »  All LLMs  »  Qwen  »  Qwen1.5 14B   URL Share it on

  20673   
  Arxiv:2309.16609   Autotrain compatible   Conversational   En   Endpoints compatible   Pretrained   Qwen2   Region:us   Safetensors   Sharded   Tensorflow
Model Card on HF 🤗: https://huggingface.co/Qwen/Qwen1.5-14B 

Qwen1.5 14B Benchmarks

MMLU Pro: 29.37
GPQA: 5.93
MUSR: 10.46
BBH: 30.06
IFEval: 29.05 vs 88 (so35)-67%
ARC: 56.57 vs 96.7 (so35)-41.5%
HellaSwag: 81.08 vs 95.3 (gpt4)-14.9%
MMLU: 69.36 vs 88.3 (so35)-21.4%
TruthfulQA: 52.06 vs 59 (gpt4)-11.8%
WinoGrande: 73.48 vs 87.5 (gpt4)-16%
GSM8K: 67.63 vs 96.4 (so35)-29.8%
MATH Lvl 5: 18.2
LLME Score: 0.30825
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
Qwen1.5 14B (Qwen/Qwen1.5-14B)

Qwen1.5 14B Parameters and Internals

Model Type 
decoder-only, transformer-based, pretrained
Use Cases 
Considerations:
We advise not using base language models for text generation without post-training.
Additional Notes 
Qwen1.5 is the beta version of Qwen2, supporting multilingual capabilities and significant performance improvements.
Supported Languages 
en (unknown)
Training Details 
Model Architecture:
Based on Transformer architecture with SwiGLU activation, attention QKV bias, group query attention, mixture of sliding window attention and full attention, etc.
LLM NameQwen1.5 14B
Repository 🤗https://huggingface.co/Qwen/Qwen1.5-14B 
Model Size14b
Required VRAM28.3 GB
Updated2024-12-21
MaintainerQwen
Model Typeqwen2
Model Files  3.9 GB: 1-of-8   4.0 GB: 2-of-8   3.9 GB: 3-of-8   3.9 GB: 4-of-8   3.9 GB: 5-of-8   4.0 GB: 6-of-8   3.1 GB: 7-of-8   1.6 GB: 8-of-8
Supported Languagesen
Model ArchitectureQwen2ForCausalLM
Licenseother
Context Length32768
Model Max Length32768
Transformers Version4.37.0
Tokenizer ClassQwen2Tokenizer
Padding Token<|endoftext|>
Vocabulary Size152064
Torch Data Typebfloat16
Errorsreplace

Best Alternatives to Qwen1.5 14B

Best Alternatives
Context / RAM
Downloads
Likes
Virtuoso Small128K / 29.6 GB196139
SuperNova Medius128K / 29.7 GB1782196
Qwen2.5 14B128K / 29.6 GB3646647
Rombos LLM V2.6 Qwen 14B128K / 29.7 GB495747
Qwen2.5 14B Wernickev3128K / 29.7 GB741
SeQwence 14Bv1128K / 29.7 GB3752
Qwen2.5 14B Emerged128K / 29.7 GB360
Lamarck 14B V0.3128K / 29.7 GB1502
QwenSlerp6 14B128K / 29.7 GB1291
...rtuosoSmall InstructModelStock128K / 29.7 GB51
Note: green Score (e.g. "73.2") means that the model is better than Qwen/Qwen1.5-14B.

Rank the Qwen1.5 14B 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? 40013 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