What are the hardware requirements for SAM GPTQ?

SAM GPTQ requires approximately 4.2 GB of VRAM and supports a context window of 32K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed SAM GPTQ and how large is it?

SAM GPTQ is developed by TheBloke, a model with 7.2b 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 SAM GPTQ?

SAM GPTQ is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

SAM GPTQ by TheBloke — VRAM 4.2GB, 32K context

Name: SAM GPTQ
Author: TheBloke

SAM GPTQ is an open-source language model by TheBloke. Features: 7.2b LLM, VRAM: 4.2GB, Context: 32K, License: apache-2.0, Quantized, LLM Explorer Score: 0.1.

4-bit Base model:quantized:superagi/... Base model:superagi/sam En Gptq Mistral Quantized Region:us Safetensors

Model Card on HF 🤗: https://huggingface.co/TheBloke/SAM-GPTQ

SAM GPTQ Parameters and Internals

Model Type

mistral

Use Cases

Areas:

reasoning

Applications:

multi-hop reasoning

Primary Use Cases:

task breakdown, improved reasoning

Limitations:

Not suitable for conversations, simple Q&A

Considerations:

Better reasoning using high-quality data, not suitable for production due to lack of moderation mechanisms.

Additional Notes

SAM is a demonstration of advanced reasoning using less but high-quality data, not suitable for production due to lack of guardrails.

Supported Languages

en (High)

Training Details

Data Sources:

OpenSource LLMs

Data Volume:

97% smaller dataset compared to other baseline models

Methodology:

Fine-tuned using a smaller, high-quality dataset

Training Time:

4 hours

Hardware Used:

NVIDIA 6 x H100 SxM (80GB)

Model Architecture:

Mistral 7B

Input Output

Input Format:

[INST] {prompt} [/INST]

Accepted Modalities:

text

Output Format:

textual response

Performance Tips:

Suggested temperature for optimal performance is 0.3

LLM Name	SAM GPTQ
Repository 🤗	https://huggingface.co/TheBloke/SAM-GPTQ
Model Name	SAM
Model Creator	SuperAGI
Base Model(s)	SAM SuperAGI/SAM
Model Size	7.2b
Required VRAM	4.2 GB
Updated	2026-07-17
Maintainer	TheBloke
Model Type	mistral
Model Files	4.2 GB
Supported Languages	en
GPTQ Quantization	Yes
Quantization Type	gptq
Model Architecture	MistralForCausalLM
License	apache-2.0
Context Length	32768
Model Max Length	32768
Transformers Version	4.35.2
Tokenizer Class	LlamaTokenizer
Vocabulary Size	32000
Torch Data Type	float16

Rank the SAM GPTQ 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 SAM GPTQ

Best Alternatives	Context / RAM	Downloads	Likes
Multi Verse Model GPTQ	32K / 4.2 GB	111	1
Garrulus GPTQ	32K / 4.2 GB	28	3
HamSter 0.1 GPTQ	32K / 4.2 GB	7	2
Phoenix GPTQ	32K / 4.2 GB	12	1
Mistral Ft Optimized 1227 GPTQ	32K / 4.2 GB	19	2
...hat 3.5 1210 Seraph Slerp GPTQ	32K / 4.2 GB	10	2
...h Openchat 3.5 1210 Slerp GPTQ	32K / 4.2 GB	6	1
Metis 0.5 GPTQ	32K / 4.2 GB	6	1
XDAN L1 Chat RL V1 GPTQ	32K / 4.2 GB	15	3
Apricot Wildflower 20 GPTQ	32K / 4.2 GB	12	1

Note: green Score (e.g. "73.2") means that the model is better than TheBloke/SAM-GPTQ.