What are the hardware requirements for Llama 3.2 3B Reflect?

Llama 3.2 3B Reflect requires approximately 6.5 GB of VRAM. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Llama 3.2 3B Reflect and how large is it?

Llama 3.2 3B Reflect is developed by emonHasan01, a model with 3b parameters. The model is published as open weights on Hugging Face and indexed on LLM Explorer with full benchmark history.

How does Llama 3.2 3B Reflect perform on standard benchmarks?

Llama 3.2 3B Reflect has the following published scores: LMSYS ELO 1167. Compare against reference models on this page or on the LLM Explorer leaderboards.

Llama 3.2 3B Reflect by emonHasan01 — VRAM 6.5GB

Name: Llama 3.2 3B Reflect
Author: emonHasan01

Llama 3.2 3B Reflect is an open-source language model by emonHasan01. Features: 3b LLM, VRAM: 6.5GB, License: llama3.2, Quantized, Instruction-Based, LLM Explorer Score: 0.44, ELO: 1167.

Arxiv:2110.14168 Base model:finetune:unsloth/ll... Base model:unsloth/llama-3.2-3... Chain-of-thought Conversational Dataset:openai/gsm8k Dataset:servicenow-ai/r1-disti... Distillation En Endpoints compatible Gguf Instruct Llama Llama-3.2 Lora Model-index Peft Quantized Reasoning Region:us Safetensors Sharded Tensorflow Unsloth

Model Card on HF 🤗: https://huggingface.co/emonHasan01/llama-3.2-3b-reflect

Llama 3.2 3B Reflect Benchmarks

LMSys ELO 1167 vs 1272 (so35) -8.3%

^nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").

What is the LLM Explorer Rank (Score)

Llama 3.2 3B Reflect Parameters and Internals

LLM Name	Llama 3.2 3B Reflect
Repository 🤗	https://huggingface.co/emonHasan01/llama-3.2-3b-reflect
Base Model(s)	unsloth/Llama-3.2-3B-Instruct unsloth/Llama-3.2-3B-Instruct
Model Size	3b
Required VRAM	6.5 GB
Updated	2026-10-19
Maintainer	emonHasan01
Instruction-Based	Yes
Model Files	0.1 GB 5.0 GB: 1-of-2 1.5 GB: 2-of-2
Supported Languages	en
GGUF Quantization	Yes
Quantization Type	gguf
Model Architecture	AutoModelForCausalLM
License	llama3.2
Model Max Length	131072
Is Biased	none
Tokenizer Class	TokenizersBackend
Padding Token	<\|finetune_right_pad_id\|>
PEFT Type	LORA
LoRA Model	Yes
PEFT Target Modules	gate_proj\|down_proj\|o_proj\|k_proj\|up_proj\|v_proj\|q_proj
LoRA Alpha	16
LoRA Dropout	0
R Param	16

Rank the Llama 3.2 3B Reflect capabilities

Have you tried this model? Rate its performance — this feedback helps the ML community find the right model for their needs.

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

Best Alternatives to Llama 3.2 3B Reflect

Best Alternatives	Context / RAM	Downloads	Likes
Kiel 2.1 Sol	0K / 0.1 GB	1859	2
...ma 3.2 3B It Ecommerce ChatBot	0K / 6.5 GB	336	8
SQL Llama3.2 3b Lora Model	0K / 0.1 GB	5	0
Kripalini AI 3B Instruct	0K / 0 GB	88	1
Jamba2 3B Turkish	0K / 6.1 GB	284	1