# Drex 2.0

> Drex 2.0 is a small decision model from Nace.AI with a 131K-token context window, built to make decisions rather than generate text. Given a state and typed questions (yes/no, multiple choice, or score), it returns a probability for every option in a single forward pass, making it a fit for routing, classification, tool selection, and policy checks that need a scored answer instead of prose.

Type: LLM · State: Ready

model path: `accounts/fireworks/models/drex-v2-0`

[Deploy on demand](https://app.fireworks.ai/dashboard/deployments/create?baseModel=accounts/fireworks/models/drex-v2-0)

Drex 2.0 is a small decision model from Nace.AI with a 131K-token context window, built to make decisions rather than generate text. Given a state and typed questions (yes/no, multiple choice, or score), it returns a probability for every option in a single forward pass, making it a fit for routing, classification, tool selection, and policy checks that need a scored answer instead of prose.

## Drex 2.0 API Features

- **On-demand Deployment**: On-demand deployments allow you to use Drex v2.0 on dedicated GPUs with Fireworks' high-performance serving stack with high reliability and no rate limits. · [Docs](https://docs.fireworks.ai/guides/ondemand-deployments)

## Metadata

- **State**: Ready
- **Created on**: 12 OCT, 2026
- **Kind**: Base model
- **Provider**: N/A
- **Huggingface**: [NaceXFireworks/Drex-v2.0](https://huggingface.co/NaceXFireworks/Drex-v2.0)

## Specifications

- **Calibrated**: No
- **Mixture-of-Experts**: No
- **Parameters**: 34.6B

## Supporting Functionality

- **Fine-tuning**: Not Supported
- **Serverless**: Not Supported
- **Context Length**: 262k tokens
- **Function Calling**: Not Supported
- **Embeddings**: Not Supported
- **Rerankers**: Not Supported
- **Supporting image input**: Not Supported
