Five additions to the AWS catalog
AWS announced on November 11 that DeepSeek-V4-Pro, DeepSeek-V4-Flash and DeepSeek-V4.1-Flash are available through Amazon SageMaker JumpStart. A separate announcement adds DeepReinforce AI’s Ornith-1.0-9B and Ornith-1.0-35B coding models.
Both AWS notices are dated November 11 and concern availability through the JumpStart deployment catalog.
Different intended workloads
AWS positions DeepSeek-V4-Pro for demanding reasoning and text generation, V4-Flash for more cost-sensitive language workloads, and V4.1-Flash for agentic and multimodal tasks. Its announcement lists a one-million-token context window for each DeepSeek model.
The Ornith announcement focuses on agentic coding. AWS’s performance and hardware-efficiency descriptions are vendor claims; Briefpixel has not benchmarked these models or verified that a particular user’s machine can run them.
Check the model card and deployment costs
AWS directs customers to the JumpStart model catalog in the SageMaker console or the SageMaker Python SDK. The guide says model cards expose the actions available for each model; fine-tuning, customisation and evaluation are not available for every catalog entry.
The pricing page says JumpStart itself has no additional charge for models or solutions, while the underlying training and inference instance hours are charged as if those resources had been created manually. Catalog access does not establish a free hosted endpoint. Review the selected model’s license and deployment configuration before launching resources.
