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

Qwen2.5 1.5B by Qwen

 »  All LLMs  »  Qwen  »  Qwen2.5 1.5B   URL Share it on

  914961   

Qwen2.5 1.5B is an open-source language model by Qwen. Features: 1.5b LLM, VRAM: 3.1GB, Context: 128K, License: apache-2.0, LLM Explorer Score: 0.31.

  Arxiv:2407.10671   Conversational   Deploy:azure   En   Endpoints compatible   Qwen2   Region:us   Safetensors
Model Card on HF 🤗: https://huggingface.co/Qwen/Qwen2.5-1.5B 

Qwen2.5 1.5B Benchmarks

MMLU Pro: 20.61
GPQA: 4.7
MUSR: 5.27
BBH: 16.66
IFEval: 26.74 vs 88 (so35)-69.6%
MATH Lvl 5: 9.14
LLME Score: 0.30791
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
Qwen2.5 1.5B (Qwen/Qwen2.5-1.5B)
🌟 Advertise your project 🚀

Qwen2.5 1.5B Parameters and Internals

Model Type 
causal language model
Use Cases 
Limitations:
Not recommended for conversations without post-training
Additional Notes 
The base 0.5B Qwen2.5 model requires post-training for certain use cases such as conversations.
Supported Languages 
en (Proficient), zh (Proficient), fr (Proficient), es (Proficient), pt (Proficient), de (Proficient), it (Proficient), ru (Proficient), ja (Proficient), ko (Proficient), vi (Proficient), th (Proficient), ar (Proficient)
Training Details 
Methodology:
Pretraining
Context Length:
32768
Model Architecture:
Transformers with RoPE, SwiGLU, RMSNorm, Attention QKV bias and tied word embeddings
LLM NameQwen2.5 1.5B
Repository 🤗https://huggingface.co/Qwen/Qwen2.5-1.5B 
Model Size1.5b
Required VRAM3.1 GB
Updated2026-05-07
MaintainerQwen
Model Typeqwen2
Model Files  3.1 GB
Supported Languagesen
Model ArchitectureQwen2ForCausalLM
Licenseapache-2.0
Context Length131072
Model Max Length131072
Transformers Version4.40.1
Tokenizer ClassQwen2Tokenizer
Padding Token<|endoftext|>
Vocabulary Size151936
Torch Data Typebfloat16
Errorsreplace

Quantized Models of the Qwen2.5 1.5B

Model
Likes
Downloads
VRAM
Qwen2.5 1.5B Unsloth Bnb 4bit2149791 GB
Qwen2.5 1.5B Instruct 4bit399800 GB
Qwen2.5 1.5B Bnb 4bit145931 GB
WiNGPT Babel413471 GB
Qwen2.5 Coder 1.5B 4bit22960 GB
Qwen2.5 1.5B Instruct 8bit2821 GB
...5B Instruct Abliterated V1 F16083 GB

Best Alternatives to Qwen2.5 1.5B

Best Alternatives
Context / RAM
Downloads
Likes
ReaderLM V2500K / 3.1 GB112973783
Reader Lm 1.5B250K / 3.1 GB566608
DeepSeek R1 Distill Qwen 1.5B128K / 3.5 GB8307481465
VibeThinker 1.5B128K / 3.5 GB2025520
Ko R1 1.5B Preview128K / 7.1 GB159
AceInstruct 1.5B128K / 3.5 GB7688020
Qwen2 1.5B128K / 3.1 GB127243102
1.5B Cold Start SFT128K / 3.5 GB45030
OpenReasoning Nemotron 1.5B128K / 3.1 GB793456
DeepScaleR 1.5B Preview128K / 7.1 GB7129583
Note: green Score (e.g. "73.2") means that the model is better than Qwen/Qwen2.5-1.5B.

Rank the Qwen2.5 1.5B 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? 53570 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.
Check out Ag3ntum — our secure, self-hosted AI agent for server management.
Release v20260328a
  Support LLM Explorer