LLM EXPLORER 67,231 MODELS INDEXED

Llama 3.2 3B Instruct Math by MergeBench

By MergeBench · 282 downloads
HF Repo ↗ Copy URL

Llama 3.2 3B Instruct Math is an open-source language model by MergeBench. Features: 3b LLM, VRAM: 6.5GB, Context: 128K, Instruction-Based.

  Arxiv:1910.09700   Conversational   Endpoints compatible   Instruct   Llama   Region:us   Safetensors   Sharded   Tensorflow
Model Card on HF 🤗: https://huggingface.co/MergeBench/Llama-3.2-3B-Instruct_math 

Llama 3.2 3B Instruct Math Parameters and Internals

LLM NameLlama 3.2 3B Instruct Math
Repository 🤗https://huggingface.co/MergeBench/Llama-3.2-3B-Instruct_math 
Model Size3b
Required VRAM6.5 GB
Updated2026-11-12
MaintainerMergeBench
Model Typellama
Instruction-BasedYes
Model Files  5.0 GB: 1-of-2   1.5 GB: 2-of-2
Model ArchitectureLlamaForCausalLM
Context Length131072
Model Max Length131072
Transformers Version4.48.1
Tokenizer ClassPreTrainedTokenizerFast
Vocabulary Size128320
Torch Data Typebfloat16

Best Alternatives to Llama 3.2 3B Instruct Math

Best Alternatives
Context / RAM
Downloads
Likes
Llama 3.2 3B Instruct128K / 6.5 GB18295722368
Llama 3.2 3B Instruct128K / 6.5 GB18620796
Llama 3.2 3B ClauseLM128K / 6.4 GB3352
...ld Llama32 3B Instruct Qa 1p25128K / 6.5 GB3401
...old Llama32 3B Instruct Qa 2p5128K / 6.5 GB3341
...3B TechWriter Instruct Unsloth128K / 6.5 GB6271
...lama 3.2 3B ORPO Advertisement128K / 6.5 GB1892
...ini Math345 GroupB Llama32 End128K / 6.5 GB7550
Llama 3.2 3B Instruct128K / 6.5 GB9919
...ama 3.2 3B TechWriter Instruct128K / 6.4 GB2260
Note: green Score (e.g. "73.2") means that the model is better than MergeBench/Llama-3.2-3B-Instruct_math.