# AWS adds two Ornith coding models to SageMaker JumpStart

_Published Wednesday, November 11, 2026 at 7:14 PM EST · AI · Latest · Tier 2 — Notable_

AWS has made DeepReinforce AI's Ornith-1.0-9B and Ornith-1.0-35B coding models available through SageMaker JumpStart. Customers can deploy the MIT-licensed models through the SageMaker console or Python SDK.

AWS says the 9B-parameter model supports a 256K-token context window and single-GPU deployment. The 35B-parameter model activates a subset of its parameters for each token and, AWS says, can run on a single 24GB consumer GPU. Both use reinforcement learning to optimize the task scaffolds that guide coding work alongside the solutions they produce; the hardware and performance descriptions are AWS's claims.

## Sources

- [Recent Announcements](https://aws.amazon.com/about-aws/whats-new/2026/01/ornith-1-0-9b-ornith-1-0-35b-on-sagemaker-jumpstart/)

---
Canonical: https://techandbusiness.org/newswire/H9KgidTruGoHd7zSPusqu1
Published: 2026-11-12T00:14:33.707Z
Story chronology: 2026-11-11T23:36:00.000Z
Retrieved: 2026-11-12T01:53:08.265Z
Publisher: Tech & Business (techandbusiness.org)
