What are the hardware requirements for Apriel 1.5 15B Thinker?

Apriel 1.5 15B Thinker requires approximately 29.7 GB of VRAM. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Apriel 1.5 15B Thinker and how large is it?

Apriel 1.5 15B Thinker is developed by ServiceNow-AI, a model with 15b 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 Apriel 1.5 15B Thinker?

Apriel 1.5 15B Thinker is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Apriel 1.5 15B Thinker by ServiceNow-AI — VRAM 29.7GB

Name: Apriel 1.5 15B Thinker
Author: ServiceNow-AI

Apriel 1.5 15B Thinker is an open-source language model by ServiceNow-AI. Features: 15b LLM, VRAM: 29.7GB, License: mit, LLM Explorer Score: 0.25.

Arxiv:2510.01141 Conversational Endpoints compatible Image-text-to-text Llava Region:us Safetensors Sharded Tensorflow

Model Card on HF 🤗: https://huggingface.co/ServiceNow-AI/Apriel-1.5-15b-Thinker

Apriel 1.5 15B Thinker Parameters and Internals

LLM Name	Apriel 1.5 15B Thinker
Repository 🤗	https://huggingface.co/ServiceNow-AI/Apriel-1.5-15b-Thinker
Model Size	15b
Required VRAM	29.7 GB
Updated	2026-08-10
Maintainer	ServiceNow-AI
Model Type	llava
Model Files	5.0 GB: 1-of-7 5.0 GB: 2-of-7 4.9 GB: 3-of-7 4.9 GB: 4-of-7 4.9 GB: 5-of-7 3.7 GB: 6-of-7 1.3 GB: 7-of-7
Model Architecture	LlavaForConditionalGeneration
License	mit
Transformers Version	4.46.3
Tokenizer Class	PreTrainedTokenizerFast
Padding Token	<pad>
Torch Data Type	bfloat16

Rank the Apriel 1.5 15B Thinker 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

Quantized Models of the Apriel 1.5 15B Thinker

Model	Likes	Downloads	VRAM
...priel 1.5 15B Thinker 6bit MLX	1	34	12 GB
...priel 1.5 15B Thinker 2bit MLX	1	21	5 GB

Best Alternatives to Apriel 1.5 15B Thinker

Best Alternatives	Context / RAM	Downloads	Likes
Apriel 1.6 15B Thinker	0K / 29.7 GB	260	306
Apriel 1.5 15B Thinker	0K / 29.7 GB	14	4
...priel 1.5 15B Thinker 6bit MLX	0K / 12.6 GB	34	1
...priel 1.5 15B Thinker 2bit MLX	0K / 5.3 GB	21	1

Note: green Score (e.g. "73.2") means that the model is better than ServiceNow-AI/Apriel-1.5-15b-Thinker.