What are the hardware requirements for Magicoder S DS 6.7B GPTQ?

Magicoder S DS 6.7B GPTQ requires approximately 3.9 GB of VRAM and supports a context window of 16K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Magicoder S DS 6.7B GPTQ and how large is it?

Magicoder S DS 6.7B GPTQ is developed by TheBloke, a model with 6.7b 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 Magicoder S DS 6.7B GPTQ?

Magicoder S DS 6.7B GPTQ is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Magicoder S DS 6.7B GPTQ by TheBloke — VRAM 3.9GB, 16K context

Name: Magicoder S DS 6.7B GPTQ
Author: TheBloke

Magicoder S DS 6.7B GPTQ is an open-source language model by TheBloke. Features: 6.7b LLM, VRAM: 3.9GB, Context: 16K, License: other, Quantized, Instruction-Based, LLM Explorer Score: 0.1.

Arxiv:2305.06161 Arxiv:2312.02120 4-bit Base model:ise-uiuc/magicoder-... Base model:quantized:ise-uiuc/... Dataset:ise-uiuc/magicoder-evo... Dataset:ise-uiuc/magicoder-oss... Gptq Instruct Llama Quantized Region:us Safetensors

Model Card on HF 🤗: https://huggingface.co/TheBloke/Magicoder-S-DS-6.7B-GPTQ

Magicoder S DS 6.7B GPTQ Parameters and Internals

Model Type

deepseek, text-generation

Use Cases

Areas:

coding tasks

Limitations:

may not work well in non-coding tasks

Considerations:

Users should be aware of the model's risks, biases, and limitations.

Training Details

Data Sources:

Magicoder-OSS-Instruct-75K, Magicoder-Evol-Instruct-110K

Methodology:

Finetuned from deepseek-coder-6.7b-base

Responsible Ai Considerations

Fairness:

OSS-Instruct mitigates inherent bias of the LLM-synthesized instruction data.

Input Output

Input Format:

Prompt_Template: 'You are an exceptionally intelligent coding assistant... @@Instruction {prompt} @@Response'

LLM Name	Magicoder S DS 6.7B GPTQ
Repository 🤗	https://huggingface.co/TheBloke/Magicoder-S-DS-6.7B-GPTQ
Model Name	Magicoder S DS 6.7B
Model Creator	Intellligent Software Engineering (iSE
Base Model(s)	ise-uiuc/Magicoder-S-DS-6.7B ise-uiuc/Magicoder-S-DS-6.7B
Model Size	6.7b
Required VRAM	3.9 GB
Updated	2026-07-25
Maintainer	TheBloke
Model Type	deepseek
Instruction-Based	Yes
Model Files	3.9 GB
GPTQ Quantization	Yes
Quantization Type	gptq
Model Architecture	LlamaForCausalLM
License	other
Context Length	16384
Model Max Length	16384
Transformers Version	4.35.2
Tokenizer Class	LlamaTokenizer
Padding Token	<｜end▁of▁sentence｜>
Vocabulary Size	32256
Torch Data Type	float16

Rank the Magicoder S DS 6.7B 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 Magicoder S DS 6.7B GPTQ

Best Alternatives	Context / RAM	Downloads	Likes
...pseek Coder 6.7B Instruct GPTQ	16K / 3.9 GB	514	25
Medicine Chat GPTQ	4K / 3.9 GB	20	6
Law Chat GPTQ	4K / 3.9 GB	5	4
Finance Chat GPTQ	4K / 3.9 GB	4	3
AmberChat GPTQ	2K / 3.9 GB	1	1
...LI LlumiX 32K Instruct F16 0.2	32K / 13.5 GB	8	1
...LI LlumiX 32K Instruct F16 0.1	32K / 13.5 GB	10	0
...rpreter DS 6.7B 6.0bpw H6 EXL2	16K / 5.2 GB	5	2
...rpreter DS 6.7B 8.0bpw H8 EXL2	16K / 6.9 GB	3	2
...rpreter DS 6.7B 4.0bpw H6 EXL2	16K / 3.6 GB	1	1

Note: green Score (e.g. "73.2") means that the model is better than TheBloke/Magicoder-S-DS-6.7B-GPTQ.