What are the hardware requirements for WizardCoder Guanaco 15B V1.0 GPTQ?

WizardCoder Guanaco 15B V1.0 GPTQ requires approximately 9.2 GB of VRAM. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed WizardCoder Guanaco 15B V1.0 GPTQ and how large is it?

WizardCoder Guanaco 15B V1.0 GPTQ is developed by TheBloke, a model with 15b parameters. The model is published as open weights on Hugging Face and indexed on LLM Explorer with full benchmark history.

How does WizardCoder Guanaco 15B V1.0 GPTQ perform on standard benchmarks?

WizardCoder Guanaco 15B V1.0 GPTQ has the following published scores: MMLU 25.88. Compare against reference models on this page or on the LLM Explorer leaderboards.

WizardCoder Guanaco 15B V1.0 GPTQ by TheBloke — VRAM 9.2GB

Name: WizardCoder Guanaco 15B V1.0 GPTQ
Author: TheBloke

WizardCoder Guanaco 15B V1.0 GPTQ is an open-source language model by TheBloke. Features: 15b LLM, VRAM: 9.2GB, License: apache-2.0, Quantized, Code Generating, LLM Explorer Score: 0.12, Arc: 32.6, HellaSwag: 45.4, MMLU: 25.9, GSM8K: 2.9.

4-bit Codegen Dataset:guanaco En Gpt bigcode Gptq Quantized Region:us Safetensors

Model Card on HF 🤗: https://huggingface.co/TheBloke/WizardCoder-Guanaco-15B-V1.0-GPTQ

WizardCoder Guanaco 15B V1.0 GPTQ Parameters and Internals

Model Type

text generation

Additional Notes

Model heavily optimized for English text generation with an emphasis on code-related instructions.

Supported Languages

English (high proficiency)

Training Details

Data Sources:

openassistant-guanaco dataset

Methodology:

finetuning with QLORA

Input Output

Input Format:

text

Accepted Modalities:

text

Output Format:

text

Performance Tips:

Ensure proper formatting of input prompts.

LLM Name	WizardCoder Guanaco 15B V1.0 GPTQ
Repository 🤗	https://huggingface.co/TheBloke/WizardCoder-Guanaco-15B-V1.0-GPTQ
Base Model(s)	LoupGarou/WizardCoder-Guanaco-15B-V1.1 LoupGarou/WizardCoder-Guanaco-15B-V1.1
Model Size	15b
Required VRAM	9.2 GB
Updated	2026-07-28
Maintainer	TheBloke
Model Type	gpt_bigcode
Model Files	9.2 GB
Supported Languages	en
GPTQ Quantization	Yes
Quantization Type	gptq
Generates Code	Yes
Model Architecture	GPTBigCodeForCausalLM
License	apache-2.0
Model Max Length	2048
Transformers Version	4.30.0.dev0
Tokenizer Class	GPT2Tokenizer
Vocabulary Size	49153
Torch Data Type	float16
Activation Function	gelu

Rank the WizardCoder Guanaco 15B V1.0 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 WizardCoder Guanaco 15B V1.0 GPTQ

Best Alternatives	Context / RAM	Downloads	Likes
...ardCoder Guanaco 15B V1.1 GPTQ	0K / 9.2 GB	50	12
WizardCoder 15B 1.0 GPTQ	0K / 9.2 GB	77	176
...lus Guanaco GPT4 15B V1.0 GPTQ	0K / 9.2 GB	9	9
Minotaur 15B GPTQ	0K / 9.2 GB	6	8
Minotaur 15B 8bit	0K / 16.5 GB	8	0
WizardCoder 15B V1.0 8bit	0K / 16.5 GB	3	1
Starcoder 15b 8bit	0K / GB	17	4
Starcoder 15b 8bit	0K / GB	11	4
Starcoder 15b 4bit	0K / GB	12	2
Starcoder 15b 4bit	0K / GB	10	2

Note: green Score (e.g. "73.2") means that the model is better than TheBloke/WizardCoder-Guanaco-15B-V1.0-GPTQ.