NVIDIA Nemotron Labs Dory1.0 4B BF16
NVIDIA Nemotron Labs Dory1.0 4B BF16 is an open-weights language model from NVIDIA, released October 16, 2026. It has 4B parameters (dense). The context window is 256K tokens.
Check it against your own PC →
Can you run NVIDIA Nemotron Labs Dory1.0 4B BF16 locally?
The smallest common GPU it runs on is 6 GB of VRAM (with 32 GB of RAM): Q6_K, fully on GPU, roughly 34 tokens per second. Estimates assume 32 GB of DDR5 RAM, a 16K context and llama.cpp-style GGUF quantization. Enter your exact hardware in the live app.
| GPU memory | Verdict | Best quant | VRAM | RAM | tok/s |
|---|---|---|---|---|---|
| 6 GB VRAM | Runs on GPU | Q6_K (≈ lossless) | 4.6 GB | – | ~34 |
| 8 GB VRAM | Runs on GPU | Q6_K (≈ lossless) | 4.6 GB | – | ~48 |
| 12 GB VRAM | Runs on GPU | Q6_K (≈ lossless) | 4.6 GB | – | ~62 |
| 16 GB VRAM | Runs on GPU | Q6_K (≈ lossless) | 4.6 GB | – | ~84 |
| 24 GB VRAM | Runs on GPU | Q6_K (≈ lossless) | 4.6 GB | – | ~150 |
| 32 GB VRAM | Runs on GPU | Q6_K (≈ lossless) | 4.6 GB | – | ~250 |
Ready-made quantizations on Hugging Face: cloudnathan5/Dory1.0-4B-GGUF · nile-academy/nemotron-dory1.0-4b-gguf · DevQuasar/nvidia.NVIDIA-Nemotron-Labs-Dory1.0-4B-BF16-GGUF · netprojects/nemotron-dory1.0-4b-mlx-8bit · netprojects/nemotron-dory1.0-4b-mlx-bf16 · snsnc/NVIDIA-Nemotron-Labs-Dory1.0-4B-mlx-4bit
Benchmark results
Epoch AI has not published results for NVIDIA Nemotron Labs Dory1.0 4B BF16 yet. New models are usually evaluated within days to weeks.
Links
Hugging Face: nvidia/NVIDIA-Nemotron-Labs-Dory1.0-4B-BF16
More from NVIDIA
NVIDIA Nemotron Labs Dory1.0 4B MXFP8 · Nemotron 3.5 Lightning · Nemotron 3 Ultra · Nemotron 3 Super · Nemotron 3-Nano-30B-A3B · Nemotron-Cascade 14B · NVIDIA Isaac GR00T N1.6 · NVIDIA-Nemotron-Nano-9B-v2