Inkling (NVFP4) — VRAM Requirements
Written by Jakub Rusinowski · Last updated July 21, 2026
How much GPU VRAM you need to run Inkling Inkling (NVFP4) by Thinking Machines locally, a 1000B-parameter model. Figures are quantized weights + KV cache + framework overhead from a memory-bandwidth roofline model.
Inkling (NVFP4) needs about 605 GB VRAM at Q4_K_M.
VRAM by Quantization
| Quant | Bits/weight | Weights | Total VRAM |
|---|---|---|---|
| Q2_K | 2.63 | 328.8 GB | 329.6 GB |
| Q3_K_M | 3.41 | 426.3 GB | 427.1 GB |
| Q4_K_M | 4.83 | 603.8 GB | 604.5 GB |
| Q5_K_M | 5.67 | 708.8 GB | 709.5 GB |
| Q6_K | 6.56 | 820.0 GB | 820.8 GB |
| Q8_0 | 8.50 | 1062.5 GB | 1063.3 GB |
| F16 | 16.00 | 2000.0 GB | 2000.8 GB |
Switch quantization in the interactive calculator, or see the full Inkling model page.
Add this badge to your model card
Model creators: paste this into your Hugging Face model card README to link readers straight to this VRAM breakdown.
[](https://llmconfigurator.com/en/vram-calculator/inkling-nvfp4)Estimates only — actual VRAM varies with context length, batch size, runtime and KV-cache settings.