# Launch Week - Day 2: Chandra is Faster (Again)

> Announcing Chandra Small, a latency-optimized model that achieves 2-3x faster speeds with minimal performance degradation.

- Canonical: https://www.datalab.to/blog/launch-week-chandra-is-faster-again
- Published: 2025-12-02
- Authors: Zach Nussbaum

TL;DR; We trained a latency-optimzed small model that is 2-3x faster with minimal performance degradations.

## Chandra Small

We recently shipped a few updates to Chandra and our API ([Making Chandra 3x faster](/blog/speeding-up-chandra) and [Chandra 1.1](/blog/launch-week-chandra-1-1)). Now, we're excited to announce Chandra Small, a latency-optimized model found exclusively in the [Datalab API](https://documentation.datalab.to/docs/welcome/api).

Shortly after our launch of [Chandra](/blog/introducing-chandra), we trained and deployed Chandra Small. In our testing, Chandra Small is **2-3x** faster than Chandra with minimal performance degradations. Additionally, we trained Chandra Small using QAT to enable quantization and further reduce latency.

![olmOCR Benchmark Scores](/images/blog/launch-week-chandra-is-faster-again/Chandra-Chart.png)

Additionally, we found that we can reduce the number of tokens needed for many pages. This led to 30% latency reductions. With Chandra Small, users can expect 2-4 pages/s on an H100.

## Try Out Chandra Small

Give [Chandra Small a try in the API](https://documentation.datalab.to/api-reference/marker#body-mode) (via `Fast` mode).

We're excited to continue to push the frontier to make Chandra as accurate and fast as possible! Reach out to [support@datalab.to](mailto:support@datalab.to) for more information or for access to a self-hosted version of Chandra Small!