FLUX.1 Dev.
Open-weight FLUX.1 — non-commercial license.
1× AMD Radeon RX 5700 XT.
Most-aggressive quantisation we have a working recommendation for. Lower precision = less VRAM = cheaper hardware, at a small accuracy cost.
Cheapest hosted endpoints.
| Provider | Access | $/M in | $/M out | |
|---|---|---|---|---|
| Replicate | hosted inference | — | — | |
| fal.ai | hosted inference | — | — |
Variants in the FLUX family.
Frequently asked.
How do I run FLUX.1 Dev?
Where can I access FLUX.1 Dev?
How much does it cost to run FLUX.1 Dev?
Is FLUX.1 Dev open-source or proprietary?
Cheapest hardware per quantisation.
Each row is one quantisation tier (the same weights compressed differently). Lower precision → lower VRAM → cheaper hardware, at the cost of small accuracy loss. $/hr refreshed hourly from each provider's API.
| Quantisation | Cheapest GPU config | Total VRAM | Live $/hr | tokens/sec | |
|---|---|---|---|---|---|
|
FP16
FP16 — half precision (default)
|
Sweet spot for FLUX.1 Dev.
|
24 GB | — | — | Compare → |
|
FP8
FP8 — 8-bit float (Hopper / Blackwell)
|
16 GB | — | — | Compare → | |
|
INT4
INT4 — 4-bit integer (~4× VRAM saving)
|
8 GB | — | — | Compare → | |
|
NF4
NF4 — 4-bit normalised
|
Halves VRAM, ~95% quality.
|
24 GB | — | — | Compare → |
What it costs per month across providers.
Estimate your monthly bill for FLUX.1 Dev across every host that publishes per-token pricing. Slide your token volumes; the chart + table re-rank cheapest-first.
No priced API access rows on file for FLUX.1 Dev yet.
Rent the GPU instead of paying per token.
For an open-weights model like FLUX.1 Dev, you can rent a GPU and serve inference yourself. The math: cheapest GPU rental × 730 hours/month + your electricity rate × power draw.
Assumes the GPU runs 24/7 at ~85% utilisation. If your traffic is bursty, you'll pay less for the API and probably more for the GPU (idle hours still cost rental). The breakeven analysis lives on the Self-host vs API breakeven tool.
About FLUX.1 Dev.
FLUX.1 Dev is the open-weight sibling of FLUX.1 Pro — distilled to be smaller while keeping most of Pro's image quality. Released under the FLUX.1 Non-Commercial license: free for research, hobbyist, and personal use; commercial deployment requires a separate license from Black Forest Labs. Runs on a single 24 GB GPU (RTX 4090 / RTX 5090).