NVIDIA's Alpamayo 1.5 10B is a chain-of-thought reasoning VLA model, exceling at interactive and steerable reasoning for autonomous-driving scenarios, particularly handling rare events. It is built on the Cosmos-Reason2 VLM backbone with a diffusion-based trajectory decoder and features 10.5 billion parameters.
Input
Output
Context
-
Max Output
1K
Parameters
10B
Input Modalities
Output Modalities
Estimates based on INT8 quantization. Actual requirements vary by framework and configuration.
Data sourced from official provider APIs and documentation
Last updated: Jun 23, 2026
Automatically route workloads to the right model for every task, every time.