Back to News
DeepGroveAugust 4, 2026

Maple-Preview

Open Source
reasoning
ternary
mixture-of-experts
on-device

Explore Maple-Preview

Visit the official website to learn more and get started

### TL;DR

Maple-Preview is an open-source 20B-A1B ternary-weight reasoning large language model optimized for efficient on-device inference. It features a 24-layer, 256-expert architecture designed to deliver state-of-the-art reasoning capabilities while maintaining high performance on consumer hardware like the Mac mini M4.

Key Insights & Metrics

Pricing
Free (Open Source)
Cost structure
Version
Preview
Current release version
Hardware
Compatible CUDA environment for Transformers implementation; separate on-device runtime for Apple Silicon
Compute requirements
Category
Open Source
Licensing model
Region
United States
Primary region

Key Features

  • 20B-A1B ternary-weight architecture
  • 256-expert configuration with 8 active experts
  • 131,072 token context window
  • High-speed inference (200+ tokens/sec on M4 Mac mini)
  • Optimized for on-device deployment

Discussion

0
Upvotes
0
Downvotes
0 reviews

Sign in to leave a review

Reviews

No reviews yet. Be the first to review!

🚀 Join the AI dev community — follow us everywhere

© 2026 MARKTECHPOST AI MEDIA INC. All rights reserved.Terms & ConditionsPrivacy Policy
Beta Mode