Deca

AI will touch every problem worth solving. Deca builds frontier models that are just as capable as they are efficient.

Scroll to explore the future of agentic AI.

Introducing Deca 2.5

Deca 2.5

Introducing Deca 2.5

Deca 2.5 is our sharpest model family yet. Use it in , with the API, or try it in

Deca Chat

Deca 2.5 is our sharpest model family yet.

Tool Use

Deca natively hooks into web browsing, terminal, and file systems to accomplish complex multi-step workflows.

Fast & Efficient

Engineered from the ground up to minimize latency and token costs without sacrificing reasoning quality.

API Access

Integrate Deca into your own products. Fully compatible with OpenAI standards for drop-in replacement.

Meet the Models

Choose the perfect model for your needs. From everyday tasks to complex research.

Deca 2.5 Mini

deca-2.5-mini

Small but intelligent model for everyday tasks.

Input
$1.00 /MTok
Output
$4.00 /MTok
Cached input
$0.15 /MTok
Cache write
$0.50 /MTok
Context: 160K tokens

Deca 2.5 Pro

deca-2.5-pro

For power users who need more intelligence for less cost.

Input
$2.00 /MTok
Output
$8.00 /MTok
Cached input
$0.30 /MTok
Cache write
$1.00 /MTok
Context: 160K tokens

Deca 2.5 Ultra

deca-2.5-ultra

Our most capable model for tasks requiring extreme intelligence.

Input
$4.00 /MTok
Output
$16.00 /MTok
Cached input
$0.60 /MTok
Cache write
$2.00 /MTok
Context: 1M tokens
Prompt caching is enabled automatically for supported requests. Cached input is billed at the cached input rate, while cache writes use the cache write rate. Send {"cache_control":{"type":"none"}} to disable caching for a request.

At Deca, we build AI that does more with less.

The Deca Manifesto Our view on what AI should be, and why the industry is missing the mark.
Mar 2026
Deca