Skip to content
NEWSee what changed →

nvidia · nemotron_h

NVIDIA-Nemotron-3.5-Lightning-30B-A3B-NVFP4

NVIDIA-Nemotron-3.5-Lightning-30B-A3B-NVFP4 by nvidia: a 17.8B-parameter mixture-of-experts open-weight model under the other license.

otherMixture of experts1049K contextAuto-imported · not yet reviewed

908,246 downloads · 504 likes · Model card · synced 2026-11-19

Specifications

Parameters
32.9B (1.5B active)
Architecture
moe · nemotron_h
Layers
52
Hidden size
2,688
KV heads · head dim
2 · 128
Vocabulary
131,072
Native dtype
—
Context window
1,048,576 tokens
Released
2026-08-04
Last modified on Hub
2026-09-10

Features & licensing

License
other
Commercial use
No
Modalities
text
Tool / function calling
No
Reasoning mode
No
Languages
EN, ES, FR, DE, IT, JA
Quantised variants on Hub
GGUF
Pipeline
text-generation
transformerssafetensorsnemotron_htext-generationnvidiapytorchnemotron-3.5conversationalenesfrdeitjaeval-resultsendpoints_compatible8-bitmodelopt

Task fit

Editorial scores (0–100) used by the recommendation engine.

  • Chat70
  • RAG / Q&A70
  • Code70
  • Summarisation70
  • Extraction70
  • Agents70

Hardware to run NVIDIA-Nemotron-3.5-Lightning-30B-A3B-NVFP4

Weights need about 72 GB at FP16, 38 GB at INT8 and 22 GB at INT4. The KV cache adds roughly 53 MB per 1,000 tokens per request. For a reference workload of 5 requests per second with 1,500 input and 300 output tokens, total GPU memory is around 28 GB, which fits on 1 × NVIDIA A100 (80 GB).

GPU configurations at the reference workload
ConfigurationVRAMUtilisationEst. first tokenCloud / month
2 × NVIDIA L4 (24 GB)48 GB58 %~0.5 s$1,168
1 × NVIDIA L40S (48 GB)48 GB58 %~0.2 s$1,387
1 × NVIDIA A100 (80 GB)recommended80 GB35 %~0.1 s$2,336
1 × NVIDIA H100 (80 GB)80 GB35 %~0.0 s$3,285

Estimates only. Run an assessment for your own traffic.

Auto-imported from Hugging Face. Architecture, license and feature data are synced from the model repository; the task-fit scores below are provisional defaults until a curator reviews them.