Nemotron340B params4k context
Nemotron 340B locally
Nemotron 340B is an open-weight LLM from the Nemotron family with 340B parameters. Main use: chat, RAG and general assistance. Detected minimum hardware: Instinct MI325X (256 GB).
Technical facts
Parameters340B
Max context4k
Q4_K_M213.7 GB
Q5_K_M261.2 GB
Q8380.0 GB
FP16760.0 GB
FamilyNemotron
Last sync2026-05-12
Available quantizations
GGUF weightsQ4_K_M
213.7GB
Acceptable. Good compromise when VRAM is limited.
Q5_K_M
261.2GB
Good quality. Sweet spot for size and precision.
Q8
380.0GB
Near-FP16 quality. Comfortable for production.
FP16
760.0GB
Reference precision. Maximum quality, doubled VRAM.
Compatible GPUs
3 single-GPUGPUs that can run Nemotron 340B on a single card, ranked by VRAM headroom.
Similar models
VRAM estimates: parameters x bits/8 plus margin. Real performance varies by engine, context length and batch size.
sync: 2026-05-12