Just Released: NVIDIA TensorRT-LLM 0.13.0

Updates include tensor parallel support for Mamba2, sparse mixer normalization for MoE models, and more.

Updates include tensor parallel support for Mamba2, sparse mixer normalization for MoE models, and more.

Source

Leave a Reply

Your email address will not be published.

Previous post My Time at Evershine has passed $1.7M in crowdfunding and its developer wants it to become ‘a must play series in the cozy/simulation RPG space’
Next post David Hayter is cranking up Metal Gear Solid fans again, teasing ‘a role I’ve not played since…’