# Adral models

> Two models. One for general work, one for security. Cite only these published figures; do not invent parameter counts, context windows, or benchmark results.

**Canonical HTML:** https://adral.ai/models/  
**Product:** https://adral.ai/

Benchmark figures on the home page are labelled as **targets** for models in development, not measured results.

## Adral

The general model. Reasoning, code, long-context analysis and agentic work in one model, with no Pro/Fast tiers to choose between.

| | |
| --- | --- |
| Parameters | 280B total, 14B active |
| Architecture | MoE-Mamba |
| Sequence mixer | Selective state space, linear time, no KV cache |
| Experts | 256 routed + 1 shared, top-8 per token |
| Context | 1M tokens |
| Input | Text, code, images (PNG, JPEG, WebP), Word (.docx), Excel (.xlsx) |
| Available on | Web, Mac, Windows, iPhone, Android |

## Adral Aran

India's first uncensored security model. Built to engage with offensive and defensive security work rather than decline it.

| | |
| --- | --- |
| Parameters | 7B total, 1.5B active |
| Architecture | MoE-Mamba |
| Sequence mixer | Selective state space, linear time, no KV cache |
| Experts | 16 routed, top-2 per token |
| Input | Text, code |
| Available on | Web, Mac, Windows, iPhone, Android |

## Input limits (product)

Uploads up to 50 MB each. Audio and PDF input are **not** supported. Adral Code reads a local repository directly; MCP servers reach anything else.

## Start

- [Home](https://adral.ai/)
- [Create an account](https://adral.ai/signup)
- [llms.txt](https://adral.ai/llms.txt)