The real cost to run Qwen3 8B NVFP4

nvidia · 4.7B parameters · 41K context · commercial OK

Qwen3 8B NVFP4 — 4.7B params (nvidia). Auto-indexed from the Hugging Face Hub (97,673 downloads). Parameter count is exact; download size and quantizations are estimates.

What it costs to run Qwen3 8B NVFP4 — $0 markup

On your own machine — $0. Runs free locally if you have about 7 GB of VRAM at Q4_K_M (point Spanvero at LM Studio, Ollama, or llama.cpp).
On your own rented GPU — from $0.26/hr. NVIDIA RTX 3090 24GB at the direct Vast.ai price (≈ $0.06 a session), $0 markup.
Via your own API key — $0.14/1M tokens (blended input + output). No GPU to manage (rough estimate for this size).

Key facts

Parameters	4.7B
Context window	41K tokens
Recommended quant	Q4_K_M
VRAM to run	~7 GB (at Q4_K_M, 16.4K context)
Download size	~3 GB
License	Commercial use OK

Open the free Spanvero advisor → for the live, interactive math for your exact workload and hardware.

Related models

Qwen3.6 35B A3B NVFP4 — 18.7B, nvidia
Gemma 4 31B IT NVFP4 — 20.9B, nvidia
NVIDIA Nemotron 3 Super 120B A12B NVFP4 — 67.2B, nvidia
NVIDIA Nemotron 3 Nano 30B A3B BF16 — 31.6B, nvidia
Gemma 4 26B A4B NVFP4 — 14.4B, nvidia

Browse: More NVIDIA models · All models · Compare