What are the hardware requirements for ALMA 13B Pretrain AWQ?

ALMA 13B Pretrain AWQ requires approximately 7.2 GB of VRAM and supports a context window of 4K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed ALMA 13B Pretrain AWQ and how large is it?

ALMA 13B Pretrain 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 ALMA 13B Pretrain AWQ?

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

ALMA 13B Pretrain AWQ by TheBloke — VRAM 7.2GB, 4K context

Name: ALMA 13B Pretrain AWQ
Author: TheBloke

ALMA 13B Pretrain AWQ is an open-source language model by TheBloke. Features: 13b LLM, VRAM: 7.2GB, Context: 4K, License: mit, Quantized, LLM Explorer Score: 0.09.

Arxiv:2309.11674 4-bit Awq Base model:haoranxu/alma-13b-p... Base model:quantized:haoranxu/... Llama Quantized Region:us Safetensors

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

ALMA 13B Pretrain AWQ Parameters and Internals

Model Type

llama

Use Cases

Primary Use Cases:

Translation

Additional Notes

Quantized by TheBloke using AWQ method. AWQ models support faster inference and are efficient for deployment with smaller GPUs.

Supported Languages

Chinese (high), English (high)

Training Details

Data Sources:

monolingual data, parallel data

Data Volume:

12B monolingual tokens

Methodology:

Full-weight and LoRA fine-tuning

Input Output

Input Format:

Translate this from Chinese to English: Chinese: {prompt} English:

Accepted Modalities:

text

Output Format:

text

LLM Name	ALMA 13B Pretrain AWQ
Repository 🤗	https://huggingface.co/TheBloke/ALMA-13B-Pretrain-AWQ
Model Name	ALMA 13B Pretrain
Model Creator	haoranxu
Base Model(s)	ALMA 13B Pretrain haoranxu/ALMA-13B-Pretrain
Model Size	13b
Required VRAM	7.2 GB
Updated	2026-07-21
Maintainer	TheBloke
Model Type	llama
Model Files	7.2 GB
AWQ Quantization	Yes
Quantization Type	awq
Model Architecture	LlamaForCausalLM
License	mit
Context Length	4096
Model Max Length	4096
Transformers Version	4.30.0.dev0
Tokenizer Class	LlamaTokenizer
Beginning of Sentence Token	<s>
End of Sentence Token	</s>
Unk Token	<unk>
Vocabulary Size	32000
Torch Data Type	float32

Rank the ALMA 13B Pretrain 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 ALMA 13B Pretrain AWQ

Best Alternatives	Context / RAM	Downloads	Likes
Yarn Llama 2 13B 128K AWQ	128K / 7.2 GB	4	2
LongAlign 13B 64K AWQ	64K / 7.2 GB	4	2
...oboros L2 13B 2 1 YaRN 64K AWQ	64K / 7.2 GB	5	2
OrcaMaid V3 13B 32K AWQ	32K / 7.2 GB	5	4
OrcaMaid V2 FIX 13B 32K AWQ	32K / 7.2 GB	7	1
NexusRaven V2 13B AWQ	16K / 7.2 GB	11	3
NexusRaven V2 13B AWQ	16K / 7.2 GB	1	3
...th CodeLlama 13B Python Hf AWQ	16K / 7.5 GB	6	0
WhiteRabbitNeo 13B AWQ	16K / 7.2 GB	332	4
NexusRaven V2 13B AWQ	16K / 7.2 GB	10	1

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