Nemotron-H models
3 Nemotron-H models from NVIDIA on Hugging Face, from 8.1B to 56.3B parameters, published by NVIDIA in BF16. The smallest official model, Nemotron-H-8B-Base-8K, needs about 18.1 GB of VRAM at its published precision; the cheapest live fit is RTX A5000 at $0.176/hr.
Part of the Nemotron series
Pick a size
One row per official Nemotron-H size: the VRAM it needs at each precision, the cheapest GPU that holds it today, and the KV cache for a 32K-token context.
| Model | Parameters | Native VRAM | FP8 VRAM | INT4 VRAM | Live GPU fit (native) | Est. $/hr | KV cache at 32K |
|---|---|---|---|---|---|---|---|
| Nemotron-H-8B-Base-8K | 8.1B | 18.1 GB | 9.1 GB | 4.5 GB | RTX A5000 | $0.176/hr | not published for this architecture |
| Nemotron-H-56B-Base-8K | 56.3B | 126 GB | 62.9 GB | 31.5 GB | AMD MI300X | $2.63/hr | not published for this architecture |
VRAM is weights times a flat 1.2 overhead; the KV cache is a separate per-model figure at 16-bit, shown where the architecture is published. See the methodology.
Official models (3)
| Model | Parameters | Published as | VRAM needed | Live GPU fit | Est. $/hr |
|---|---|---|---|---|---|
| Nemotron-H-8B-Base-8K | 8.1B | BF16 | 18.1 GB | RTX A5000 | $0.176/hr |
| Nemotron-H-8B-Reasoning-128K | 8.1B | BF16 | 18.1 GB | RTX A5000 | $0.176/hr |
| Nemotron-H-56B-Base-8K | 56.3B | BF16 | 126 GB | AMD MI300X | $2.63/hr |
VRAM is for the precision the model is published in: the weights times a flat 1.2 overhead, with the KV cache not included. See the methodology. The fit shown is the lowest-priced single GPU type that holds the model at that precision, or the lowest-priced multi-GPU set (up to 8) when none does.
More from NVIDIA
- Parakeet (8 models)
- Nemotron 3 (19 models)
- Cosmos (7 models)
- Nemotron Nano (6 models)
- nvidia (2 models)
- Nemotron Labs (5 models)
- Nemotron Speech (1 models)
- Nemotron Parse (1 models)
- Nemotron Cascade (1 models)
- Canary (2 models)
- Nemotron VoiceChat (1 models)