GLM-5.2 FP8, NVIDIA-Nemotron-Nano-12B-v2 and GLM-OCR models now available on Amazon SageMaker JumpStart

This article announces the availability of three new foundation models on Amazon SageMaker JumpStart: GLM-5.2 FP8, NVIDIA-Nemotron-Nano-12B-v2, and GLM-OCR.

• GLM-5.2 FP8 optimized for long-horizon agentic engineering with 1M-token context window for full-cycle software development
• NVIDIA-Nemotron-Nano-12B-v2 delivers hybrid Mamba-2 and Transformer architecture with 128K context and 6x higher inference throughput
• GLM-OCR provides accurate document understanding for PDFs, handwritten notes, tables, and multilingual text with low latency
• Deploy models directly from SageMaker JumpStart console or Python SDK with minimal configuration

These models expand AWS customers' access to specialized foundation models for agentic workflows, efficient reasoning, and advanced document processing on SageMaker infrastructure.