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

Qwen2 57B A14B by Qwen

 »  All LLMs  »  Qwen  »  Qwen2 57B A14B   URL Share it on

  5150   
  Autotrain compatible   Conversational   En   Endpoints compatible   Moe   Pretrained   Qwen2 moe   Region:us   Safetensors   Sharded   Tensorflow
Model Card on HF 🤗: https://huggingface.co/Qwen/Qwen2-57B-A14B 

Qwen2 57B A14B Benchmarks

MMLU Pro: 43.51
GPQA: 7.49
MUSR: 10.54
BBH: 38.88
IFEval: 31.13 vs 88 (so35)-64.6%
MATH Lvl 5: 18.66
LLME Score: 0.30512
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
Qwen2 57B A14B (Qwen/Qwen2-57B-A14B)

Qwen2 57B A14B Parameters and Internals

Model Type 
text-generation
Supported Languages 
language_support_level (multiple), languages (English, Multilingual)
Training Details 
Model Architecture:
Transformer with SwiGLU activation, attention QKV bias, group query attention
LLM NameQwen2 57B A14B
Repository 🤗https://huggingface.co/Qwen/Qwen2-57B-A14B 
Model Size57b
Required VRAM115.2 GB
Updated2025-02-22
MaintainerQwen
Model Typeqwen2_moe
Model Files  4.0 GB: 1-of-29   4.0 GB: 2-of-29   4.0 GB: 3-of-29   4.0 GB: 4-of-29   4.0 GB: 5-of-29   4.0 GB: 6-of-29   4.0 GB: 7-of-29   4.0 GB: 8-of-29   4.0 GB: 9-of-29   4.0 GB: 10-of-29   4.0 GB: 11-of-29   4.0 GB: 12-of-29   4.0 GB: 13-of-29   4.0 GB: 14-of-29   4.0 GB: 15-of-29   4.0 GB: 16-of-29   4.0 GB: 17-of-29   4.0 GB: 18-of-29   4.0 GB: 19-of-29   4.0 GB: 20-of-29   4.0 GB: 21-of-29   4.0 GB: 22-of-29   4.0 GB: 23-of-29   4.0 GB: 24-of-29   4.0 GB: 25-of-29   4.0 GB: 26-of-29   4.0 GB: 27-of-29   4.0 GB: 28-of-29   3.2 GB: 29-of-29
Supported Languagesen
Model ArchitectureQwen2MoeForCausalLM
Licenseapache-2.0
Context Length131072
Model Max Length131072
Transformers Version4.40.1
Tokenizer ClassQwen2Tokenizer
Padding Token<|endoftext|>
Vocabulary Size151936
Torch Data Typebfloat16
Errorsreplace

Best Alternatives to Qwen2 57B A14B

Best Alternatives
Context / RAM
Downloads
Likes
Quill 57B Tokenizer128K /  GB150
Qwen2 57B A14B Instruct32K / 115.2 GB509279
Qwen2 57B A14B Instruct 8bit32K / 61.1 GB121
Qwen2 57B A14B Instruct 4bit32K / 32.3 GB82
...n2 57B A14B Instruct GPTQ Int432K / 31.6 GB14222
...7B A14B Instruct GPTQ Int4 MLX32K / 32.3 GB471
Note: green Score (e.g. "73.2") means that the model is better than Qwen/Qwen2-57B-A14B.

Rank the Qwen2 57B A14B 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