What are the hardware requirements for CodeLlama 13B AWQ?

CodeLlama 13B AWQ requires approximately 7.2 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 CodeLlama 13B AWQ and how large is it?

CodeLlama 13B AWQ is developed by TheBloke, a model with 13b 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 CodeLlama 13B AWQ?

CodeLlama 13B AWQ is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

CodeLlama 13B AWQ by TheBloke — VRAM 7.2GB, 16K context

Name: CodeLlama 13B AWQ
Author: TheBloke

CodeLlama 13B AWQ is an open-source language model by TheBloke. Features: 13b LLM, VRAM: 7.2GB, Context: 16K, License: llama2, Quantized, Code Generating, LLM Explorer Score: 0.09.

Arxiv:2308.12950 4-bit Awq Base model:codellama/codellama... Base model:quantized:codellama... Code Codegen Llama Llama2 Quantized Region:us Safetensors

Model Card on HF 🤗: https://huggingface.co/TheBloke/CodeLlama-13B-AWQ

CodeLlama 13B AWQ Parameters and Internals

Model Type

text-generation, code synthesis

Use Cases

Areas:

commercial, research

Applications:

code synthesis, code understanding

Primary Use Cases:

general code generation

Limitations:

Use only in English, Violations of applicable laws

Considerations:

Perform safety testing tailored to specific applications.

Supported Languages

English (general proficiency)

Training Details

Data Sources:

Llama 2 data with different weights

Data Volume:

See Section 2 and Table 1 in the research paper

Methodology:

Auto-regressive transformer architecture

Context Length:

4096

Training Time:

400K GPU hours of computation

Hardware Used:

A100-80GB GPUs

Model Architecture:

Optimized transformer architecture

Safety Evaluation

Methodologies:

English testing

Findings:

Model may produce inaccurate or objectionable responses

Risk Categories:

misinformation, objectionable content

Ethical Considerations:

Responsible Use Guide is available.

Responsible Ai Considerations

Fairness:

Testing in English, broader scenarios not covered.

Transparency:

Model details are public via research paper.

Accountability:

Developers should perform safety testing tailored to specific applications.

Mitigation Strategies:

Strong emphasis on safety evaluations and responsible use.

Input Output

Input Format:

Text

Accepted Modalities:

text

Output Format:

Text

Performance Tips:

Utilize optimized transformer architecture for faster inference.

LLM Name	CodeLlama 13B AWQ
Repository 🤗	https://huggingface.co/TheBloke/CodeLlama-13B-AWQ
Model Name	CodeLlama 13B
Model Creator	Meta
Base Model(s)	CodeLlama 13B Hf codellama/CodeLlama-13b-hf
Model Size	13b
Required VRAM	7.2 GB
Updated	2026-07-14
Maintainer	TheBloke
Model Type	llama
Model Files	7.2 GB
Supported Languages	code
AWQ Quantization	Yes
Quantization Type	awq
Generates Code	Yes
Model Architecture	LlamaForCausalLM
License	llama2
Context Length	16384
Model Max Length	16384
Transformers Version	4.32.0.dev0
Tokenizer Class	CodeLlamaTokenizer
Beginning of Sentence Token	<s>
End of Sentence Token	</s>
Unk Token	<unk>
Vocabulary Size	32016
Torch Data Type	bfloat16

Rank the CodeLlama 13B AWQ 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 CodeLlama 13B AWQ

Best Alternatives	Context / RAM	Downloads	Likes
...th CodeLlama 13B Python Hf AWQ	16K / 7.5 GB	6	0
Ramgpt 13B AWQ Gemm	16K / 7.2 GB	0	1
CodeLlama 13B Instruct AWQ	16K / 7.2 GB	261	9
NexusRaven 13B AWQ	16K / 7.2 GB	22	4
MAmmoTH Coder 13B AWQ	16K / 7.2 GB	6	1
...odeLlama 13B Oasst Sft V10 AWQ	16K / 7.2 GB	7	1
CodeLlama 13B Python AWQ	16K / 7.2 GB	14	2
...ma 13B Instruct Hf W4 G128 AWQ	16K / 7.2 GB	10	0
...lama 13B Python Hf W4 G128 AWQ	16K / 7.2 GB	5	0
WhiteRabbitNeo 13B V1	16K / 26 GB	3146	453

Note: green Score (e.g. "73.2") means that the model is better than TheBloke/CodeLlama-13B-AWQ.