Flux
Modes: image · video · Models: 8
Vendor: Black Forest Labs · Official API docs: docs.bfl.ml
Flux (by Black Forest Labs) is a family of high-quality text-to-image models with crisp detail and strong prompt adherence. The Kontext models add reference-guided editing — generate from a prompt and one or more input images. Flux 3 Video extends the family to motion: text-to-video and image-to-video up to 20 seconds at FHD, with native audio.
Models
| id | Name | Input type | Notes |
|---|---|---|---|
flux-2-pro | Flux 2 Pro | t2i | Multi-image input, up to 2K |
flux-2-max | Flux 2 Max | t2i | Highest quality |
flux-2-flex | Flux 2 Flex | t2i | Flexible / cost-aware |
flux-kontext-max | Flux Kontext Max | t2i | Reference-guided editing |
flux-kontext-pro | Flux Kontext Pro | t2i | Reference-guided editing |
flux-3-video | Flux 3 Video | t2v | Text-to-video with native audio |
flux-video-upscale | Flux Video Upscale | v2v | Upscales an existing video up to 3x |
flux-video-edit | FLUX Video Edit | v2v | Edit videos with a text instruction — change objects, styles or scenes while preserving motion, timing and audio. Source clips up to 15 seconds; output at 24 fps, up to 720p. |
CLI
# text-to-image
gen-ai generate -m flux-2-pro -p "studio shot of a ceramic cup, soft light" --ar 4:3
# multi-image reference (Kontext)
gen-ai generate -m flux-kontext-pro -p "place the product on a marble table" -i ./product.pngMCP
{ "name": "picsart_generate",
"arguments": {
"model": "flux-2-pro",
"prompt": "studio shot of a ceramic cup, soft light",
"aspectRatio": "4:3",
"count": 1
} }Parameters
Full parameter surface for every model, sourced from gen-ai models info <id> --json. CLI flags show the primary short form; the canonical --kebab-case long form always works too.
flux-2-pro — Flux 2 Pro
Try flux-2-pro in Playground ↗
Input type: t2i
| Param | CLI flag | Type | Values |
|---|---|---|---|
prompt | -p | text | required |
aspectRatio | --ar | enum | 1:1 · 5:3 · 3:5 · 4:3 · 3:4 (default 4:3) |
count | -n | enum | 1 · 2 · 4 · 6 · 8 · 10 (default 1) |
imageUrls | -i | file | image (up to 4) |
flux-2-max — Flux 2 Max
Try flux-2-max in Playground ↗
Input type: t2i
| Param | CLI flag | Type | Values |
|---|---|---|---|
prompt | -p | text | required |
aspectRatio | --ar | enum | 1:1 · 5:3 · 3:5 · 4:3 · 3:4 (default 1:1) |
count | -n | enum | 1 · 2 · 4 · 6 · 8 · 10 (default 1) |
imageUrls | -i | file | image (up to 1) |
flux-2-flex — Flux 2 Flex
Try flux-2-flex in Playground ↗
Input type: t2i
| Param | CLI flag | Type | Values |
|---|---|---|---|
prompt | -p | text | required |
aspectRatio | --ar | enum | 1:1 · 5:3 · 3:5 · 4:3 · 3:4 (default 3:4) |
count | -n | enum | 1 · 2 · 4 · 6 · 8 · 10 (default 1) |
imageUrls | -i | file | image (up to 1) |
flux-kontext-max — Flux Kontext Max
Try flux-kontext-max in Playground ↗
Input type: t2i
| Param | CLI flag | Type | Values |
|---|---|---|---|
prompt | -p | text | required |
aspectRatio | --ar | enum | 1:1 · 16:9 · 9:16 · 4:3 · 3:4 · 21:9 · 9:21 (default 1:1) |
count | -n | enum | 1 · 2 · 4 · 6 · 8 · 10 (default 1) |
imageUrls | -i | file | image (up to 4) |
flux-kontext-pro — Flux Kontext Pro
Try flux-kontext-pro in Playground ↗
Input type: t2i
| Param | CLI flag | Type | Values |
|---|---|---|---|
prompt | -p | text | required |
aspectRatio | --ar | enum | 1:1 · 16:9 · 9:16 · 4:3 · 3:4 · 21:9 · 9:21 (default 1:1) |
count | -n | enum | 1 · 2 · 4 · 6 · 8 · 10 (default 1) |
imageUrls | -i | file | image (up to 1) |
Notes: Flux 2 Pro and Flux Kontext Max accept up to 4 reference images; Flux 2 Max / Flex and Flux Kontext Pro accept 1. Kontext models use
imageUrlsas the edit reference.
flux-3-video — Flux 3 Video
Try flux-3-video in Playground ↗
Input type: t2v
| Param | CLI flag | Type | Values |
|---|---|---|---|
prompt | -p | text | required |
aspectRatio | --ar | enum | auto · 21:9 · 2:1 · 16:9 · 4:3 · 1:1 · 3:4 · 9:16 (default auto) |
resolution | -r | enum | hd · fhd (default hd) |
duration | -d | enum | auto · 5 · 10 · 15 · 20 (default auto) |
imageUrls | -i | file | image (up to 10) |
videoUrl | --video | file | video |
generateAudio | --audio-gen | boolean | true · false (default true) |
safetyTolerance | --safety-tolerance | range | 0–4 (default 2) |
draft | --draft | boolean | true · false (default false) |
Notes:
flux-3-videoacceptsimageUrls(image-to-video, up to 10 references) orvideoUrl, so it coverst2v,i2v, andv2vfrom one id. Draft mode is HD-only — with--drafton,fhdis rejected.
flux-video-upscale — Flux Video Upscale
Try flux-video-upscale in Playground ↗
Input type: v2v
| Param | CLI flag | Type | Values |
|---|---|---|---|
videoUrl | --video | file | required video |
upscaleFactor | --upscale-factor | range | 1.5–3, step 0.5 (default 2) |
creativity | --creativity | enum | 0 (Precise) · 1 (Creative) (default 1) |
prompt | -p | text | free text |
safetyTolerance | --safety-tolerance | range | 0–4 (default 2) |
flux-video-edit — FLUX Video Edit
Try flux-video-edit in Playground ↗
Input type: v2v
| Param | CLI flag | Type | Values |
|---|---|---|---|
videoUrl | --video | file | required video |
prompt | -p | text | required (≤4096 chars) |
safetyTolerance | --safety-tolerance | range | 0–4 (default 2) |
Pricing
gen-ai pricing flux-2-pro -n 4Cost scales with resolution/quality and the number of outputs (count).