Ornith-1.5-122B-NVFP4

Overview

Ornith-1.5-122B compressed to NVIDIA's NVFP4 - the self-improvement flagship for vLLM on Blackwell hardware, with a 262K context window and vision support.

Good for

• agentic
• chat
• coding

Recommended hardware

≈ 96 GB VRAM to run this 122B model at the NVFP4 quantization.

Needs 128 GB+ of VRAM or unified memory (e.g. multi-GPU with 2x 96 GB cards, or a 128 GB Mac)

Links

• View on HuggingFace
• Run on Infersec