What are the hardware requirements for Mamba GPT 3B V2?

Mamba GPT 3B V2 requires approximately 6.8 GB of VRAM. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Mamba GPT 3B V2 and how large is it?

Mamba GPT 3B V2 is developed by CobraMamba, a model with 3b parameters. The model is published as open weights on Hugging Face and indexed on LLM Explorer with full benchmark history.

Where can I download or evaluate Mamba GPT 3B V2?

Mamba GPT 3B V2 is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Mamba GPT 3B V2 by CobraMamba — VRAM 6.8GB

Name: Mamba GPT 3B V2
Author: CobraMamba

Mamba GPT 3B V2 is an open-source language model by CobraMamba. Features: 3b LLM, VRAM: 6.8GB, License: apache-2.0, LLM Explorer Score: 0.08.

En Gpt Llama Lora Pytorch Region:us Safetensors

Model Card on HF 🤗: https://huggingface.co/CobraMamba/mamba-gpt-3b-v2

Mamba GPT 3B V2 Parameters and Internals

Model Type

Large Language Model, Text Generation

Use Cases

Areas:

Research, Commercial Applications

Additional Notes

The model uses LLaMA architecture for causal language modeling.

Training Details

Methodology:

Fine-tuned model surpassing the original on several evaluations.

Model Architecture:

LLaMA (LlamaForCausalLM) with 32 layers of LlamaDecoderLayer each consisting of LlamaAttention (uses Linear, RotaryEmbedding) and LlamaMLP (uses Linear, SiLUActivation).

Input Output

Input Format:

<|prompt|>Your input~~<|answer|>

Output Format:

Generated text corresponding to the prompt.

Performance Tips:

Ensure appropriate setup for transformer, accelerate, and torch.

LLM Name	Mamba GPT 3B V2
Repository 🤗	https://huggingface.co/CobraMamba/mamba-gpt-3b-v2
Model Size	3b
Required VRAM	6.8 GB
Updated	2026-08-25
Maintainer	CobraMamba
Model Files	6.8 GB 6.8 GB
Supported Languages	en
Model Architecture	AutoModelForCausalLM
License	apache-2.0
Model Max Length	2048
Is Biased	none
Tokenizer Class	LlamaTokenizer
Beginning of Sentence Token	<s>
End of Sentence Token	</s>
Unk Token	<unk>
PEFT Type	LORA
LoRA Model	Yes
PEFT Target Modules	q_proj\|v_proj
LoRA Alpha	16
LoRA Dropout	0.1
R Param	256

Rank the Mamba GPT 3B V2 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 Mamba GPT 3B V2

Best Alternatives	Context / RAM	Downloads	Likes
Granite 3B Mup	4K / 14 GB	374	0
Llama 3.2 3B Mathdaily Chatbot	0K / 19.3 GB	9	0
Llama 3.2 3B Mathdaily Chatbot	0K / 6.5 GB	5	0
MM Alpaca 3B Lora	0K / 0.2 GB	12	0
Qwen2.5 3b Lora Model	0K / 0.1 GB	8	0
SQL Llama3.2 3b Lora Model	0K / 0.1 GB	5	0
...ma 3.2 3B It Ecommerce ChatBot	0K / 6.5 GB	336	8

Note: green Score (e.g. "73.2") means that the model is better than CobraMamba/mamba-gpt-3b-v2.