towsty/aigen
1
0
Fork 0
Code Issues Pull Requests 1 Actions Packages Projects Releases Wiki Activity
258 Commits 2 Branches 0 Tags
09c96c0ae7214b5f54f240f2578c228d9fe85a58
Commit Graph
19 Commits
Author SHA1 Message Date
Towsty 53bb58bbd5 Add optional reference photo conditioning to Flux generation 2026-09-06 23:05:36 -05:00
Towsty 666c922f40 Add per-image iteration settings and steps/CFG sweeps 2026-09-06 17:02:42 -05:00
towsty 2435167a41 Add Image Iterate: nested multi-prompt still families. 
Iterate runs a series of prompts as generate (no still), edit (still A), or compose (A+B), saves each output under one library collection like video extensions, and defaults to iterating from the original still.
 2026-09-04 22:46:19 -05:00
TowstyandCursor ce85e7f513 Join pre/prompt/post into the first MiniMax prompt, not only later shots. 
The motion prompt field is what Comfy actually runs. Wrappers are now composed into that string on submit, and shot 1 falls back to it if the separate pre/post fields do not arrive.

Co-authored-by: Cursor <cursoragent@cursor.com>
 2026-09-03 07:24:00 -05:00
TowstyandCursor 1206afe8fd Keep family pre/post text on shot queues and wrap the mid prompt at Comfy time. 
Later shots were starting a new job without those wrappers, so only the first clip got them.

Co-authored-by: Cursor <cursoragent@cursor.com>
 2026-09-03 07:17:00 -05:00
TowstyandCursor 9fd96cb990 Apply and print SNOFS from the LoRA dials, not a leftover Concept LoRA row. 
Co-authored-by: Cursor <cursoragent@cursor.com>
 2026-09-02 21:33:30 -05:00
TowstyandCursor 369eb595d7 Join pre/prompt/post for Comfy, stop zero-size generates snapping to a small square, and give the phone library a real closeable preview. 
Co-authored-by: Cursor <cursoragent@cursor.com>
 2026-09-02 20:29:44 -05:00
TowstyandCursor f788fd4bb4 Split prompts into pre/prompt/post and tidy the AIGen video form. 
Keep wrappers off the library mid-prompt so Use as input and Extend restore the same split, send stills to xAIGen as hidden inputs, and stop the two-column library from growing with the list.

Co-authored-by: Cursor <cursoragent@cursor.com>
 2026-09-01 18:39:02 -05:00
TowstyandCursor 755aaeede8 Overhaul Studio controls and add image batch from one still. 
Video and Image pickers, thumbwheels, aspect icons, and a Scale to MP paddle; extra prompts default to batch so each run starts from the original image.

Co-authored-by: Cursor <cursoragent@cursor.com>
 2026-08-30 23:16:19 -05:00
TowstyandCursor e89b47b5f9 Add same-mode extra passes to Image v2, and hide Shot script from image Recommend. 
Library tiles show a check when selected, and xAIGen can hide a preview from the details pane.

Co-authored-by: Cursor <cursoragent@cursor.com>
 2026-08-30 18:17:17 -05:00
TowstyandCursor 0e1222d035 Name outputs from the input, engine, and LoRA keys instead of aigen_0001. Image-edit presets now store prompts, and aspect fill pads stills without stretching. 
Co-authored-by: Cursor <cursoragent@cursor.com>
 2026-08-30 16:24:49 -05:00
TowstyandCursor bfe94e4bb1 Add an xAIGen SNOFS switch for Flux and Krea. 
It is the same LoRA card as the rest of the chain; the toggle only adds or removes that card so it no longer comes back after you turn it off.

Co-authored-by: Cursor <cursoragent@cursor.com>
 2026-08-30 13:27:30 -05:00
TowstyandCursor cdc69d2bd6 Let Image v2 pick a LoRA or LoRA chain. 
Co-authored-by: Cursor <cursoragent@cursor.com>
 2026-08-29 17:28:32 -05:00
TowstyandCursor 5fe91e6bae Show Flux and Krea as Image v2 engine cards, and run Krea on edit, compose, and refine. 
Krea image-to-image uses encode plus denoise graphs, not Klein reference latents, and still fails closed if the Krea models are missing.

Co-authored-by: Cursor <cursoragent@cursor.com>
 2026-08-29 07:29:31 -05:00
TowstyandCursor ab5477c999 Keep the Klein Concept LoRA on xAIGen only. 
AIGen hides the sliders, ignores snofs strengths, and strips the SNOFS node so that file cannot be queued from the public studio.

Co-authored-by: Cursor <cursoragent@cursor.com>
 2026-08-29 07:08:33 -05:00
TowstyandCursor f6663444f5 Add Krea as a second Image v2 Generate engine beside Flux. 
Generate can queue krea_v2_generate.json without sharing Klein encoders or falling back if Krea assets are missing.

Co-authored-by: Cursor <cursoragent@cursor.com>
 2026-08-29 07:05:01 -05:00
TowstyandCursor b35f9af10d Add Image v2 Generate for Klein text-to-image. 
New sibling graph and mode so T2I does not borrow Edit, Compose, or Refine. Stills are ignored; size is width x height only.

Co-authored-by: Cursor <cursoragent@cursor.com>
 2026-08-28 22:56:33 -05:00
TowstyandCursor a2895b8054 Add Image v2 Refine for on-device regional edits. 
Mask painter and denoise strength sit beside Edit and Compose. The canned hand/chest prompt helper is gone so that text stays yours.

Co-authored-by: Cursor <cursoragent@cursor.com>
 2026-08-28 22:03:20 -05:00
TowstyandCursor 21fe2412f9 Add Image v2 as a sibling of Image edit, with separate Klein graphs. 
Keep the v1 Image tab on its dual-branch template. The host agent follows Desktop's live port so Coolify stays on :8198.

Co-authored-by: Cursor <cursoragent@cursor.com>
 2026-08-28 21:22:37 -05:00