Llama-3.2-1B-Instruct

Meta · 1.2B · ctx 128K · llama3.2 · HuggingFace

Llama-3.2-1B-Instruct (Meta) is a 1.2B parameters open-weight model tracked on AI Hashrate. At 8K context we estimate roughly 1.9 GB VRAM for Q4 (weights 0.66 GB + KV 0.24 GB + 1 GB overhead) and 3.64 GB for FP16. About 50 GPUs/accelerators in our catalog fully fit this model at Q4 under that context assumption. Fastest Q4 configs: B200 192GB SXM ≈ 4242.4 tok/s (estimated); MI300X 192GB ≈ 2810.6 tok/s (estimated); H200 141GB SXM5 ≈ 2545.5 tok/s (estimated); H100 80GB SXM5 ≈ 1776.5 tok/s (estimated); Gaudi 3 128GB ≈ 1737.3 tok/s (estimated). Use the table below for VRAM fit and tok/s per dollar (list MSRP). Estimates follow memory-bandwidth math; measured rows override when present. See methodology for details. Methodology.

100 hardware configs — sorted by tok/s. Default fit context 8K.