Home / Companies / HuggingFace / Blog / Post Details
Content Deep Dive

Apriel-1.6-15b-Thinker: Cost-efficient Frontier Multimodal Performance

Blog post from HuggingFace

Post Details
Company
Date Published
Author
Sathwik Tejaswi Madhusudhan, Sagar Davasam, and Torsten Scholak
Word Count
1,908
Language
-
Hacker News Points
-
Summary

Apriel-1.6-15b-Thinker, a new multimodal reasoning model in ServiceNow's Apriel SLM series, achieves state-of-the-art performance with 15 billion parameters, rivaling models ten times its size, while maintaining cost-efficiency and reducing token usage by over 30%. Trained on NVIDIA DGX™ Cloud with GB200 Grace™ Blackwell Superchips, it excels in text and vision reasoning, outperforming previous versions like Apriel-1.5-15b-Thinker and competitors such as Gemini 2.5 Flash and Claude Haiku 4.5. The training process involved a depth-upscaling phase, continual pretraining, supervised fine-tuning, and reinforcement learning, focusing on reasoning quality and token efficiency. Apriel-1.6 is evaluated across various benchmarks, demonstrating strong performance in domains like tool use, math, coding, and visual reasoning, while maintaining a low compute footprint. Despite some limitations in handling complex visual tasks, its development showcases how strategic data use and training methodologies can produce a powerful model with limited resources, making it a practical choice for enterprise applications.