Zentra

Development

0%

Skip to content
Zentra
All Models
Coming Sooncodereasoninganalysislong-context

Zentra 14B

A powerful model for demanding applications requiring deep reasoning, complex code generation, and nuanced understanding.

14.2B params
131,072 tokens context
Transformer (Decoder-only)
Coming Q4 2026

Technical Specifications

Parameters14.2 billion
Context131,072 tokens
ArchitectureTransformer
Layers40
Heads40
Hidden Size5,120
Vocab128,000
LicenseApache 2.0

System Requirements

Minimum
RAM32 GB
VRAM16 GB
OSLinux, macOS, Windows
Recommended
RAM64 GB
VRAM24 GB
OSLinux, macOS

About

Zentra 14B is built for serious workloads. With 14 billion parameters and an extended context window, it handles complex tasks that smaller models struggle with.

Ideal for: - Large codebase analysis and refactoring - Complex mathematical proofs and scientific reasoning - Long-document analysis and synthesis - High-quality creative writing and translation

Get Zentra 14B

Download the GGUF weights for local inference, or try it in the playground.

Open SourceApache 2.0 LicenseHugging Face