The real cost to run Qwen3 Coder 30B A3B Instruct FP4

NVFP4 · 15.6B parameters · 262.1K context · commercial OK

Qwen3 Coder 30B A3B Instruct FP4 — 15.6B params (NVFP4). Auto-indexed from the Hugging Face Hub (11,191 downloads). Parameter count is exact; download size and quantizations are estimates.

What it costs to run Qwen3 Coder 30B A3B Instruct FP4 — $0 markup

• On your own machine — $0. Runs free locally if you have about 12 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 (≈ $0.18 a session), $0 markup.
• Via your own API key — $0.22/1M tokens (blended input + output). No GPU to manage (rough estimate for this size).

Key facts

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

Related models

• Qwen3 30B A3B NVFP4 — 15.6B, nvidia
• DeepSeek-Coder-V2-Lite Instruct — 15.7B, DeepSeek
• DeepSeek V2 Lite Chat — 15.7B, deepseek-ai
• DeepSeek V2 Lite — 15.7B, deepseek-ai
• starcoder — 15.8B, bigcode

Browse: All models · Compare

Spanvero · What's new · Prices as of 2026-06-22. We're an honest advisor — $0 markup, your own accounts, we never resell compute. © 2026 Cynosure LLC.