TEXT TO IMAGE, OR EDIT WITH REFERENCES
Start from a text prompt, or run image to image with up to 4 reference images per generation. Models with edit endpoints route automatically. No mode juggling.
IMAGE GENERATOR
THE WORLD'S BEST
IMAGE MODELS BUILT IN
Nano Banana Pro, GPT Image 2, FLUX.2, Seedream, and more. Plug the outputs into the Beeble pipeline.
01/ MODELS
EVERY FRONTIER IMAGE MODEL
ONE CREDIT POOL
Google, OpenAI, Black Forest Labs, ByteDance, Recraft, xAI, and more.
Picked from one model menu, billed from one balance.
- GOOGLE
Nano Banana Pro · 2 · 2 Lite
1K – 4K
- OPENAI
GPT Image 2 · 2 High
1K – 4K
- BLACK FOREST LABS
FLUX.2 Pro · Max · Flex · Klein
1K – 2K
- BYTEDANCE
Seedream 5.0 Pro · V5 Lite · V4.5
Up to 4K
- RECRAFT
Recraft V4.1 Pro · Utility · Utility Pro
1K – 2K
- Z-IMAGE
Z-Image Base · Turbo
1K – 2K
- XAI
Grok Imagine · Imagine Quality
1K – 2K
- + MORE
ALWAYS CURRENT
New frontier models land as they ship
02/ CAPABILITIES
TEXT TO IMAGE, IMAGE TO IMAGE
UP TO 4K
One prompt box covers generation and editing.
Resolution, aspect ratio, and variations set per run.
1K, 2K, AND 4K OUTPUT
Pick the resolution per run, model-dependent. Flagship models render true 4K stills ready for finishing work.
10 BASE RATIOS, UP TO 14
From 21:9 to 9:16, with Gemini models extending to 14 ratios. Match Image 1 sizes the output to your reference image's exact dimensions.
1–4 VARIATIONS, EVERY RUN KEPT
Generate up to 4 variations per run. Every generation lands in a history feed with full run details and one-click prompt copy.
03/ PIPELINE
AN IMAGE GENERATOR WIRED INTO A VFX PIPELINE
Every result carries Continue in Canvas.
One click sends it into Beeble Canvas, the node-graph editor.
Generic image tools stop at the still. Beeble starts there.
PLATEFRAMEIMAGE GEN (GENERATE HANGAR SET)CAMERA ANGLESWITCHXCOMPOSITE
EVERY MODEL
ONE WORKSPACE
Sign in, pick a model, and generate.
A full VFX pipeline is one click away from every result.