Compare commits

..
Author SHA1 Message Date
Brandon Thomas f587e19d25 remove gpu loading error 2025-09-27 18:27:33 -04:00
2277 changed files with 23728 additions and 68999 deletions
-1
View File
@@ -1,7 +1,6 @@
SERVER_ENVIRONMENT=development
DATABASE_URL=mysql://storyteller:password@localhost/storyteller
#DATABASE_URL=sqlite:/tmp/tasks.sqlite
#DATABASE_URL=mysql://root:root@localhost/storyteller
#DATABASE_URL=mysql://root@localhost/storyteller
+2 -2
View File
@@ -200,10 +200,10 @@ jobs:
# Use npx to ensure we use the local nx version
echo "Building with npx nx..."
npx nx run artcraft:build
npx nx run editor3d:build
echo "Checking build output..."
ls -la apps/artcraft/dist/ | head -10
ls -la apps/editor3d/dist/ | head -10
# Update Tauri config to skip the build (since we already built)
cd ${{ github.workspace }}/crates/desktop/artcraft
@@ -75,10 +75,10 @@ jobs:
# Use npx to ensure we use the local nx version
echo "Building with npx nx..."
npx nx run artcraft:build
npx nx run editor3d:build
echo "Checking build output..."
dir apps/artcraft/dist/ | Select-Object -First 10
dir apps/editor3d/dist/ | Select-Object -First 10
# Update Tauri config to skip the build (since we already built)
cd ${{ github.workspace }}/crates/desktop/artcraft
@@ -104,4 +104,4 @@ jobs:
releaseBody: "Windows build - See the assets to download this version and install."
releaseDraft: true
prerelease: false
args: ${{ matrix.args }}
args: ${{ matrix.args }}
-3
View File
@@ -45,9 +45,6 @@ crates/lib/sqlite_queries/target/
# Javascript build directories
node_modules/
# Frontend build outputs
frontend/apps/artcraft/bundle-report.html
# Vim noise
*.swo
*.swp
@@ -1,12 +0,0 @@
{
"db_name": "SQLite",
"query": "\n UPDATE tasks\n SET is_dismissed_by_user = 1\n WHERE id = ?\n ",
"describe": {
"columns": [],
"parameters": {
"Right": 1
},
"nullable": []
},
"hash": "13d435d8f681b73e4080748dcef0579e246e41049af67e1bd79ffc06aa6ba974"
}
@@ -1,12 +0,0 @@
{
"db_name": "SQLite",
"query": "\n UPDATE tasks\n SET\n task_status = ?,\n on_complete_batch_token = ?,\n on_complete_primary_media_file_token = ?,\n on_complete_primary_media_file_class = ?,\n on_complete_primary_media_file_cdn_url = ?,\n on_complete_primary_media_file_thumbnail_url_template = ?\n WHERE\n provider = ?\n AND provider_job_id = ?\n ",
"describe": {
"columns": [],
"parameters": {
"Right": 8
},
"nullable": []
},
"hash": "1e3758851fc8451670f86a9432009b73bdb81de4a08dcaa96a86afb5f9be7a68"
}
@@ -1,12 +0,0 @@
{
"db_name": "SQLite",
"query": "\n UPDATE tasks\n SET task_status = \"cancelled_by_user\"\n WHERE task_status NOT IN (\n \"complete_success\",\n \"complete_failure\",\n \"dead\",\n \"cancelled_by_user\",\n \"cancelled_by_provider\",\n \"cancelled_by_us\"\n )\n ",
"describe": {
"columns": [],
"parameters": {
"Right": 0
},
"nullable": []
},
"hash": "30272128d83543d9fbfca930b7cdd4af30adfb4d90f2de044463a111c6340495"
}
@@ -1,68 +0,0 @@
{
"db_name": "SQLite",
"query": "\n SELECT\n id,\n task_status,\n task_type,\n model_type,\n provider,\n provider_job_id,\n frontend_caller,\n frontend_subscriber_id,\n frontend_subscriber_payload\n FROM tasks\n WHERE\n provider = ?\n AND provider_job_id = ?\n ",
"describe": {
"columns": [
{
"name": "id",
"ordinal": 0,
"type_info": "Text"
},
{
"name": "task_status",
"ordinal": 1,
"type_info": "Text"
},
{
"name": "task_type",
"ordinal": 2,
"type_info": "Text"
},
{
"name": "model_type",
"ordinal": 3,
"type_info": "Text"
},
{
"name": "provider",
"ordinal": 4,
"type_info": "Text"
},
{
"name": "provider_job_id",
"ordinal": 5,
"type_info": "Text"
},
{
"name": "frontend_caller",
"ordinal": 6,
"type_info": "Text"
},
{
"name": "frontend_subscriber_id",
"ordinal": 7,
"type_info": "Text"
},
{
"name": "frontend_subscriber_payload",
"ordinal": 8,
"type_info": "Text"
}
],
"parameters": {
"Right": 2
},
"nullable": [
false,
false,
false,
true,
true,
true,
true,
true,
true
]
},
"hash": "38ff2df237b16f077865711dd81261de1b5fa05a9787679ce9ccd4ac576b1069"
}
@@ -1,116 +0,0 @@
{
"db_name": "SQLite",
"query": "\n SELECT\n id,\n task_status,\n task_type,\n model_type,\n provider,\n provider_job_id,\n frontend_caller,\n frontend_subscriber_id,\n frontend_subscriber_payload,\n on_complete_primary_media_file_token,\n on_complete_primary_media_file_class,\n on_complete_batch_token,\n on_complete_primary_media_file_cdn_url,\n on_complete_primary_media_file_thumbnail_url_template,\n created_at as \"created_at: DateTime<Utc>\",\n updated_at as \"updated_at: DateTime<Utc>\",\n completed_at as \"completed_at: DateTime<Utc>\"\n FROM tasks\n WHERE is_dismissed_by_user == 0\n ",
"describe": {
"columns": [
{
"name": "id",
"ordinal": 0,
"type_info": "Text"
},
{
"name": "task_status",
"ordinal": 1,
"type_info": "Text"
},
{
"name": "task_type",
"ordinal": 2,
"type_info": "Text"
},
{
"name": "model_type",
"ordinal": 3,
"type_info": "Text"
},
{
"name": "provider",
"ordinal": 4,
"type_info": "Text"
},
{
"name": "provider_job_id",
"ordinal": 5,
"type_info": "Text"
},
{
"name": "frontend_caller",
"ordinal": 6,
"type_info": "Text"
},
{
"name": "frontend_subscriber_id",
"ordinal": 7,
"type_info": "Text"
},
{
"name": "frontend_subscriber_payload",
"ordinal": 8,
"type_info": "Text"
},
{
"name": "on_complete_primary_media_file_token",
"ordinal": 9,
"type_info": "Text"
},
{
"name": "on_complete_primary_media_file_class",
"ordinal": 10,
"type_info": "Text"
},
{
"name": "on_complete_batch_token",
"ordinal": 11,
"type_info": "Text"
},
{
"name": "on_complete_primary_media_file_cdn_url",
"ordinal": 12,
"type_info": "Text"
},
{
"name": "on_complete_primary_media_file_thumbnail_url_template",
"ordinal": 13,
"type_info": "Text"
},
{
"name": "created_at: DateTime<Utc>",
"ordinal": 14,
"type_info": "Int64"
},
{
"name": "updated_at: DateTime<Utc>",
"ordinal": 15,
"type_info": "Int64"
},
{
"name": "completed_at: DateTime<Utc>",
"ordinal": 16,
"type_info": "Int64"
}
],
"parameters": {
"Right": 0
},
"nullable": [
false,
false,
false,
true,
true,
true,
true,
true,
true,
true,
true,
true,
true,
true,
false,
false,
true
]
},
"hash": "47f94c127abc20683b25c257aed3c2190b9bd29eb829d61392f08e10fd986767"
}
@@ -1,12 +0,0 @@
{
"db_name": "SQLite",
"query": "\n UPDATE tasks\n SET\n task_status = ?,\n on_complete_batch_token = ?,\n on_complete_primary_media_file_token = ?,\n on_complete_primary_media_file_class = ?,\n on_complete_primary_media_file_cdn_url = ?,\n on_complete_primary_media_file_thumbnail_url_template = ?\n WHERE id = ?\n ",
"describe": {
"columns": [],
"parameters": {
"Right": 7
},
"nullable": []
},
"hash": "50a2449b87abed7df3302a74ecd0e8ab87940077a247636680cb09b2fe488321"
}
@@ -1,12 +0,0 @@
{
"db_name": "SQLite",
"query": "\n UPDATE tasks\n SET is_dismissed_by_user = 1\n WHERE is_dismissed_by_user != 1\n ",
"describe": {
"columns": [],
"parameters": {
"Right": 0
},
"nullable": []
},
"hash": "f83018ec5493982659a56576296914f937bb4e75db3a522eb4bda486d5caa402"
}
Generated
+875 -1132
View File
File diff suppressed because it is too large Load Diff
+6 -36
View File
@@ -6,19 +6,17 @@ resolver = "2"
members = [
# API clients for external services
# These might be custom implementations or wrappers over other client libraries.
"crates/api_clients/artcraft/artcraft_api_defs", # TODO: api_clients/{artcraft}/{api_defs,client}
"crates/api_clients/artcraft/storyteller_client",
"crates/api_clients/artcraft_api_defs", # TODO: api_clients/{artcraft}/{api_defs,client}
"crates/api_clients/cloud_storage",
"crates/api_clients/fal_client",
"crates/api_clients/google_drive_common",
"crates/api_clients/google_sign_in",
"crates/api_clients/grok_client",
"crates/api_clients/midjourney_client",
"crates/api_clients/openai_sora_client",
"crates/api_clients/openart_client",
"crates/api_clients/runwayml_client",
"crates/api_clients/storyteller_client",
"crates/api_clients/thumbnail_generator",
"crates/api_clients/world_labs_client",
# CLI tools for operations, etc.
"crates/cli/devenv/dev-database-seed",
@@ -45,7 +43,6 @@ members = [
"crates/lib/caching/memory_store",
"crates/lib/caching/redis_caching",
"crates/lib/cloudflare_errors",
"crates/lib/cloudflare_mitigation",
"crates/lib/concurrency",
"crates/lib/cookie_store",
"crates/lib/cookies",
@@ -68,7 +65,6 @@ members = [
"crates/lib/files/media",
"crates/lib/files/mimetypes",
"crates/lib/files/videos",
"crates/lib/http_headers",
"crates/lib/idempotency",
"crates/lib/image_algorithms",
"crates/lib/jobs_common",
@@ -88,7 +84,6 @@ members = [
"crates/lib/testing",
"crates/lib/tts_common",
"crates/lib/url_config",
"crates/lib/url_utils",
"crates/lib/user_input_common",
"crates/lib/web_scrapers",
#"crates/lib/api_clients/fakeyou_client",
@@ -186,11 +181,10 @@ workspace-build-acceleration = { path = "crates/workspace-build-acceleration" }
# Internal libs
# Setting paths here ensures we don't need to do so in every crate's Cargo.toml
# It makes refactoring much easier.
artcraft_api_defs = { path = "crates/api_clients/artcraft/artcraft_api_defs" }
artcraft_api_defs = { path = "crates/api_clients/artcraft_api_defs" }
build_metadata = { path = "crates/lib/build_metadata" }
cloud_storage = { path = "crates/api_clients/cloud_storage" }
cloudflare_errors = { path = "crates/lib/cloudflare_errors" }
cloudflare_mitigation = { path = "crates/lib/cloudflare_mitigation" }
component_traits = { path = "crates/service/plugins/component_traits" }
cookie_store = { path = "crates/lib/cookie_store" }
crockford = { path = "crates/lib/data/crockford" }
@@ -200,8 +194,6 @@ errors = { path = "crates/lib/errors" }
fal_client = { path = "crates/api_clients/fal_client" }
file_to_base64_url = { path = "crates/lib/files/file_to_base64_url" }
filesys = { path = "crates/lib/files/filesys" }
grok_client = { path = "crates/api_clients/grok_client" }
http_headers = { path = "crates/lib/http_headers" }
idempotency = { path = "crates/lib/idempotency" }
images = { path = "crates/lib/files/images" }
jwt_light = { path = "crates/lib/jwt_light" }
@@ -213,14 +205,12 @@ openai_sora_client = { path = "crates/api_clients/openai_sora_client" }
primitives = { path = "crates/lib/data/primitives" }
shared_service_components = { path = "crates/lib/shared_service_components" }
sqlite_tasks = { path = "crates/schema/database/sqlite_tasks" }
storyteller_client = { path = "crates/api_clients/artcraft/storyteller_client" }
storyteller_client = { path = "crates/api_clients/storyteller_client" }
subprocess_common = { path = "crates/lib/subprocess_common" }
test_data = { path = "crates/lib/test_data" }
testing = { path = "crates/lib/testing" }
tokens = { path = "crates/schema/public/tokens" }
url_utils = { path = "crates/lib/url_utils" }
web_base64 = { path = "crates/lib/data/web_base64" }
world_labs_client = { path = "crates/api_clients/world_labs_client" }
# External libs
# Centrally pinned versions so that we can uniformly upgrade/downgrade versions across the workspace
@@ -237,28 +227,22 @@ actix-web-actors = "4.3.1+deprecated"
actix-web-lab = "0.20.2"
anyhow = "1.0.98"
base64 = "0.22.1"
bstr = "1.12.1"
bytes = "1.10.1"
chrono = { version = "0.4.41", features = ["serde"] }
cookie = "0.18.1"
env = "1.0.1"
env_logger = "0.11.8"
futures = { version = "0.3.31" }
hex-literal = "1.0.0"
indexmap = "2.9.0"
log = "0.4.27"
once_cell = "1.21.3"
r2d2 = "0.8.10"
rand = "0.9.1"
redis = { version = "1.0.2", features = ["r2d2", "tokio-rustls-comp", "tokio-comp", "cluster-async", "aio"] }
regex = "1.12.2"
reqwest = { version = "0.12.23", default-features = false, features = [ "charset", "http2", "system-proxy", "json", "cookies", "rustls-tls" ] }
regex = "1.11.1"
reqwest = { version = "0.12.22", features = ["json", "cookies"] }
reqwest-websocket = { version = "0.5.0" }
serde = { version = "1.0.219", features = ["derive"] }
serde_derive = "1.0.219"
serde_json = "1.0.140"
serde_yaml = "0.9.34+deprecated"
sha2 = "0.11.0-rc.2"
sha3 = "0.10.8"
tauri = { version = "2.8.3", features = ["unstable", "devtools"] } # NB: "devtools" prevents App Store distribution; "ubstable" is needed for window access
tauri-plugin-dialog = "2.3.3"
@@ -271,17 +255,3 @@ thiserror = "2.0.12"
tokio = { version = "1.45.1", features = ["macros"] }
toml = "0.8.22"
url = { version = "2.5.4", features = ["serde"] }
uuid = "1.18.1"
# NB: wreq gets past fingerprinting, but we have to deal with platform-specific build issues.
#
# Mac:
# brew install --formula cmake pkg-config llvm
# Windows:
# choco install cmake -y
# choco install strawberryperl -y
# choco install pkgconfiglite -y
# choco install llvm -y
# choco install nasm -y
wreq = { version = "6.0.0-rc.20", features = ["json", "multipart", "cookies", "ws", "stream"] }
wreq-util = "3.0.0-rc.6"
-86
View File
@@ -1,86 +0,0 @@
ArtCraft License (WIP)
======================
(**Warning**: This isn't a real lawyery license full of legitimate legalese yet, but we'll take care of that ASAP.)
ArtCraft is a form of "fair source" software. It's like [free beer](https://en.wiktionary.org/wiki/free_as_in_beer)
in that we give you the code, but there are a few small limitations on what you can do with it.
First, what you *can* do with ArtCraft:
> 1. You do not have to pay us to use ArtCraft. It's free.
>
> 2. You can back up, copy, modify, compile, and use the code for your own private personal purposes.
>
> 3. You can do anything you want with the media you create with ArtCraft. Generated assets are yours.
What you *can't do* with ArtCraft:
> 1. You cannot sell the ArtCraft software commercially (eg. as your own software).
>
> 2. You cannot use the ArtCraft code to develop a competing business or product (eg. selling AI image and video tools).
>
> 3. You cannot fork ArtCraft and remove ArtCraft community, donation links, or paid model services. We intend to
> make money to improve ArtCraft for everyone.
>
> 4. You cannot use the ArtCraft name, logo, or mascot characters to promote your business without permission.
This license means you can see what we're doing and not be at risk of having the rug pulled out from under
you (eg. if our founders die, if our business collapses, if zombies or space aliens invade). It also gives
you a really great way to customize the software or lend a hand in reporting (or even fixing) bugs.
Our desire is to support artists, designers, and filmmakers. We want to help you build your own creative studios
and businesses. We want to give you creative software you can use forever.
ArtCraft is free. We do offer models available through us where you can pay us and support our development, but we also
offer the ability for you to use other providers - and we'll keep adding more options so you have maximum flexibility.
## Our Promise
- All development and business plans will be communicated openly with the community.
- Except for minor test scripts or a few small places where we might need to use licensed 3rd party code, all of
our engineering efforts will live in a single publicly available monorepo. You can see everything we do and even
benefit from using it for personal use. We plan on building a lot more than what you see today, and you'll get
all of it.
- We will not "rug pull" our users. If our business shuts down, gets acquired, or otherwise changes, the tools
we create will always continue to function in the same way.
- If the business ends and our founders quit working in this space, we'll make it so the software transitions
to an OSI-approved open source license. (We don't intend to quit!)
## Why not OSI Open Source?
Philosophical answer: The OSI is an industry body led by hyperscalers. They poach from open source all the time, yet
hyperscaler systems and code (AWS, GCP, YouTube, etc.) remain closed source. The tiniest nuggets are open, but the vast
machinery surrounding it - systems, user behaviors, funnels, at-scale moneymakers - is totally closed.
Practical answer: OSI licenses make it exceptionally challenging to grow big things commercially. And many of
the licenses leave room for hyperscalers to reach in and cherry-pick pieces for themselves that they can better
commercialize than the original authors.
Then there's completely open source software that doesn't even attempt to be a business.
We love software like Blender, but (1) developing ArtCraft is our full-time job, (2) we want to build something at scale
that is much bigger than what you presently see, and (3) this will eventually become much more capital intensive than
just writing code - we want to train novel models that work directly in conjunction with the user interface.
If and when we do get to scale, we intend to keep every bit of our work in the open, including the models. Available
for use, inspection, learning, fitting onto consumer GPU hardware, and more.
Not having a way to make money is unfortunately what killed Stability.ai. Stability.ai walked so that others could
eventually run. (Black Forest Labs, et al.) We're not egotistical enough to count ourselves in that party yet, but we
believe in the original Stability.ai ethos and that their model can work with a few tweaks.
We want to build big and incredible tools the world hasn't seen before. We want to keep them free and entirely source
available. You don't have to pay us, but if you do, all of your investment goes right back into the engineering
and the tools. We're writing code you can clone and keep.
## Inspiration
We took inspiration for this from [Fair Source](https://fair.io/), [Fair Code](https://faircode.io/),
the [Defold License](https://defold.com/license/), and Hacker News discussions. The folks at Obsidian
have some pretty good thoughts on this as well.
~ Brandon Thomas
+45 -70
View File
@@ -1,75 +1,50 @@
ArtCraft
========
ArtCraft is the IDE for interactive AI image and video creation.
We turn prompting into *crafting*, so your ideas become a form of tangible expression and computing.
This is Adobe Photoshop for everyone, and we're giving away the source code!
storyteller-rust
================
## Show, Don't Tell: Advanced Crafting Features
Our Artcraft monorepo, which serves our desktop application, backend infrastructure, and more.
Text-to-image is great, but artists *need control*. It's important to know what your image will look like before you generate it, and it's vitally important to achieve consistency and repeatability.
| Feature | Demo + Explanation
|-----------------------------------|-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
| **Image to Location** | ![Video](https://github.com/user-attachments/assets/21f103e3-cc19-4882-a630-9caa1b76ae31) Placing virtual actors into physical environments establishes single-location consistency. You can film multiple shots within a room without having things disappear. |
| **3D Image Compositing** | ![Video](https://github.com/user-attachments/assets/f93a616f-571d-474e-bcc0-53736de7303d) Use images (backdrops, foreground elements, props, etc.) in scenes with depth and blend them naturally together. Just a couple of images usually leads to great compositions. |
| **2D Image Compositing** | ![Video](https://github.com/user-attachments/assets/d6f99391-e496-4c62-9e37-29734ba5f899) Use images, background removal, layers, and simple drawing tools to precisely compose a scene. |
| **Image to 3D Mesh** | ![Video](https://github.com/user-attachments/assets/600a405c-e360-48c1-9b42-6e657ae6243b) It's almost impossible to lay out complicated objects or block complicated scenes; turning images into 3D helps position elements exactingly and intentionally. |
| **Character Posing** | ![Video](https://github.com/user-attachments/assets/52a8e983-7c8f-42d2-be8b-25296ab9ed57) You can dynamically pose your characters to achieve the precise character, scene, and camera blocking before calling "action". |
| **Scene Blocking w/ Kit Bashing** | ![Video](https://github.com/user-attachments/assets/eef025ac-0346-4a46-a023-d48e23629eb5) Use 3D asset kits to precisely block out your scene: get the correct angles, object positions, and rich depth layering you can't with text prompting. |
| **Character Identity Transfer** | ![Video](https://github.com/user-attachments/assets/629119ee-8c76-4a83-9827-8c6c995a3ec1) Use mannequins as simple 3D ControlNets for posing any character. |
| **Background Removal** | ![Video](https://github.com/user-attachments/assets/90c65057-5531-404f-83af-b34e66e24ec1) Remove backgrounds from images to make them useful in 2D or 3D compositing. They can be props, layers, or backdrops. |
| **Mixed Asset Crafting** | ![Video](https://raw.githubusercontent.com/storytold/github-media/main/ship-editing.gif) You can use image cutouts, worlds, and simple 3D meshes all together to precisely and intentionally lay out your scenes. |
| **Scene Blocking** | (preview coming soon) |
| **Canvas Editing** | (preview coming soon) |
| **Scene Relighting** | (preview coming soon) |
Note: all of the above videos were generated for free with Grok Video; the cost to build this README was negligible.
## Quick and Easy Prompting
We haven't abandoned text-to-asset generation for quick prototyping and ideation. We support every popular workflow in a first class fashion.
| Feature | Demo + Explanation
|-----------------------|-------------------------------------------------------------------------------------------------------------------------------------------------------|
| **Text to Image** | ![Video](https://github.com/user-attachments/assets/9cc289cd-faf4-4eaf-aed2-21134cce127c) Text prompt over a dozen different image models. |
| **Image Editing** | ![Video](https://github.com/user-attachments/assets/a06fa6ad-936c-42d0-8767-48fdbb8ff141) Edit with Nano Banana Pro and GPT Image 1.5. |
| **Image Editing** | ![Video](https://github.com/user-attachments/assets/f036e08a-f3a6-417a-98ee-ec7f04b2b5ff) Use inpainting, drawing, masking, etc. to edit images. |
| **Image to Video** | ![Video](https://github.com/user-attachments/assets/2bc6c592-511e-4fba-b40f-03c96699b7f7) Image to video with lots of different options and controls. |
| **Image Inpainting** | (preview coming soon) |
| **Image Ingredients** | (preview coming soon) |
## Models and Providers Supported within Artcraft
| Provider | Features |
|------------|-------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
| Artcraft | Nano Banana, Nano Banana Pro, GPT-Image-1 / 1.5, Seedream 4 / 4.5, Flux 1.1 / Kontext, Veo 2 / 3 / 3.1, Kling 1.6 / 2.1 / 2.5 / 2.6, Seedance, Sora 2 / Pro, Hunyuan 3d |
| Grok | Grok Imagine, Grok Video |
| Midjourney | Image Gen (all versions) |
| Sora | Sora 1, Sora 2, GPT-Image-1 |
| WorldLabs | Marble (Gaussian Splat World Generation) |
We're going to be adding the following providers soon: Kling (via Kling website accounts), Google (via API keys),
Runway (via website account), Luma (via website account).
We're potentially interested in adding other aggregators for those who already have subscriptions and credits at
those providers, for example: OpenArt, FreePik, etc.
## Downloads
- [Visit our website for the stable Windows and MacOS releases](https://getartcraft.com/)
- Or you can grab a [more recent Windows and MacOS build directly](https://github.com/storytold/artcraft/releases)
- Linux requires building from source for now
## Documentation
```
."-,.__
`. `. ,
.--' .._,'"-' `.
. .' `'
`. / ,'
` '--. ,-"'
`"` | \
-. \, |
`--Y.' ___.
\ L._, \
_., `. < <\ _
,' ' `, `. | \ ( `
../, `. ` | .\`. \ \_
,' ,.. . _.,' ||\l ) '".
, ,' \ ,'.-.`-._,' | . _._`.
,' / \ \ `' ' `--/ | \ / / ..\
.' / \ . |\__ - _ ,'` ` / / `.`.
| ' .. `-...-" | `-' / / . `.
| / |L__ | | / / `. `.
, / . . | | / / ` `
/ / ,. ,`._ `-_ | | _ ,-' / ` \
/ . \"`_/. `-_ \_,. ,' +-' `-' _, ..,-. \`.
. ' .-f ,' ` '. \__.---' _ .' ' \ \
' / `.' l .' / \.. ,_|/ `. ,'` L`
|' _.-""` `. \ _,' ` \ `.___`.'"`-. , | | | \
|| ,' `. `. ' _,...._ ` | `/ ' | ' .|
|| ,' `. ;.,.---' ,' `. `.. `-' .-' /_ .' ;_ ||
|| ' V / / ` | ` ,' ,' '. ! `. ||
||/ _,-------7 ' . | `-' l / `||
. | ,' .- ,' || | .-. `. .' ||
`' ,' `".' | | `. '. -.' `'
/ ,' | |,' \-.._,.'/'
. / . . \ .''
.`. | `. / :_,'.'
\ `...\ _ ,'-. .' /_.-' S T O R Y T E L L E R
`-.__ `, `' . _.>----''. _ __ / Turning memes into dreams
.' /"' | "' '_
/_|.-'\ ,". '.'`__'-( \
/ ,"'"\,' `/ `-.|"
```
- [developer documentation](./_docs)
- [server API reference](https://storyteller-docs.netlify.app/)
- [tools, scripts, misc](./script)
- [license](./LICENSE.md)
- [roadmap](./ROADMAP.md)
## Social
[![discord](https://raw.githubusercontent.com/CLorant/readme-social-icons/main/large/colored/discord.svg)](https://discord.gg/artcraft)
[![youtube](https://raw.githubusercontent.com/CLorant/readme-social-icons/main/large/colored/youtube.svg)](https://www.youtube.com/@OfficialArtCraftStudios)
[![x](https://raw.githubusercontent.com/CLorant/readme-social-icons/main/large/colored/linkedin.svg)](https://www.linkedin.com/company/artcraft-ai)
[![x](https://raw.githubusercontent.com/CLorant/readme-social-icons/main/large/colored/twitter-x.svg)](https://x.com/get_artcraft)
-47
View File
@@ -1,47 +0,0 @@
ArtCraft Roadmap
================
## 2026 Goals (non-exhaustive)
Product
- Make ArtCraft better and more useful than model aggregation websites (our #1 goal).
We don't dislike those websites; we just set a high standard for ourselves.
- Build the best "tangible" computing surfaces where you can mold scenes and videos like
clay and direct them with precision and clarity.
- Add every model and provider into ArtCraft so you can build your own crafting IDE.
Nothing will be left out.
- Build purpose-specific image models, eg. for relighting scenes and reposing characters
and props.
Studio
- Build our content studio. Make lots of narrative shorts and at least one
feature-length AI film by the year's end. Theatrical distribution, but also lots
of social media content. (We are creators, and we want to employ more creators!)
- Host festivals and competitions, including in-person meetups and screenings.
Architecture
- Remove dependence on ArtCraft hosted services.
- More Rust, more speed.
- Code cleanup, more/better tests, CI, ...
Community
- Build a safe community free of negativity and anti-AI stigma. A community of filmmakers
and artists.
## Loftier Goals
- Build an *Open* OpenAI / Google / Anthropic. Open models, open platforms, open hardware.
Why? Because the future could be bleak without it. How, you might ask? Get in touch if
you're curious.
- Don't Be Evil™
@@ -1,20 +0,0 @@
-- noinspection SqlDialectInspectionForFile
-- noinspection SqlNoDataSourceInspectionForFile
-- noinspection SqlResolveForFile
select
u.display_name,
count(*) as sessions,
a.session_duration_seconds / 60 / 60 as hours,
a.measurement_count,
u.email_address,
a.os_platform,
a.os_version,
u.created_at,
u.maybe_source,
a.last_active_at
from analytics_app_sessions as a
left outer join users as u
on a.user_token = u.token
group by a.user_token
order by hours desc;
@@ -9,14 +9,6 @@
-- tasks_v1.sqlite - initial version
-- tasks_v2.sqlite - added model_type (nullable)
-- tasks_v3.sqlite - added frontend_caller, comments
-- tasks_v4.sqlite - added is_dismissed_by_user
-- tasks_v5.sqlite - added five fields:
-- on_complete_batch_token,
-- on_complete_primary_media_file_token,
-- on_complete_primary_media_file_class,
-- on_complete_primary_media_file_cdn_url,
-- on_complete_primary_media_file_thumbnail_url_template,
--
CREATE TABLE tasks (
-- Task auto-incrementing primary key.
@@ -58,32 +50,6 @@ CREATE TABLE tasks (
-- will be re-emitted back to the frontend.
frontend_subscriber_payload TEXT,
-- Whether the user has dismissed the task from view.
is_dismissed_by_user INTEGER NOT NULL DEFAULT 0,
-- OPTIONAL.
-- When the generation completes, the batch token (if any).
-- This will point to a collection of multiple files.
on_complete_batch_token TEXT,
-- OPTIONAL.
-- When the generation completes, the media token of the "primary" media file.
-- We'll probably always generate homogenous types, but just in case this is the first or primary item in a batch.
on_complete_primary_media_file_token TEXT,
-- OPTIONAL.
-- When the generation completes, the file type of the primary media file.
-- We'll probably always generate homogenous types, but just in case this is the first or primary item in a batch.
on_complete_primary_media_file_class TEXT,
-- OPTIONAL.
-- The URL to the media file.
on_complete_primary_media_file_cdn_url TEXT,
-- OPTIONAL.
-- The URL template for the thumbnail of the primary media file.
on_complete_primary_media_file_thumbnail_url_template TEXT,
created_at INTEGER NOT NULL DEFAULT (unixepoch('now')),
updated_at INTEGER NOT NULL DEFAULT (unixepoch('now')),
completed_at INTEGER DEFAULT NULL
-1
View File
@@ -152,7 +152,6 @@ CREATE TABLE media_files (
-- * `vmd`
-- * `pmx`
-- * `csv`
-- * `spz` Gaussian splats
media_type VARCHAR(16) NOT NULL,
-- Broad class of media:
@@ -1,7 +0,0 @@
use serde_derive::{Deserialize, Serialize};
use utoipa::ToSchema;
#[derive(Serialize, Deserialize, ToSchema)]
pub struct SimpleGenericJsonSuccess {
pub success: bool,
}
@@ -1 +0,0 @@
pub mod remove_image_background;
@@ -1,5 +0,0 @@
pub mod bg_removal;
pub mod edit;
pub mod inpaint;
pub mod multi_function;
pub mod text;
@@ -1,69 +0,0 @@
use serde_derive::{Deserialize, Serialize};
use tokens::tokens::generic_inference_jobs::InferenceJobToken;
use tokens::tokens::media_files::MediaFileToken;
use utoipa::ToSchema;
pub const BYTEDANCE_SEEDREAM_V4_MULTI_FUNCTION_IMAGE_GEN_PATH: &str = "/v1/generate/image/multi_function/bytedance_seedream_4";
#[derive(Serialize, Deserialize, ToSchema)]
pub struct BytedanceSeedreamV4MultiFunctionImageGenRequest {
/// Idempotency token to prevent duplicate requests.
pub uuid_idempotency_token: String,
/// Text prompt to generate the image from.
pub prompt: Option<String>,
/// Image media tokens to include in the editing context.
/// If present, we're doing image editing (image-to-image / image-editing)
/// If absent, we're doing image generation (text-to-image)
pub image_media_tokens: Option<Vec<MediaFileToken>>,
/// Number of images to generate. Default is one.
pub num_images: Option<BytedanceSeedreamV4MultiFunctionImageGenNumImages>,
pub max_images: Option<BytedanceSeedreamV4MultiFunctionImageGenMaxImages>,
pub image_size: Option<BytedanceSeedreamV4MultiFunctionImageGenImageSize>,
}
#[derive(Copy, Clone, Debug, Serialize, Deserialize, ToSchema)]
#[serde(rename_all = "snake_case")]
pub enum BytedanceSeedreamV4MultiFunctionImageGenNumImages {
One, // Default
Two,
Three,
Four,
}
#[derive(Copy, Clone, Debug, Serialize, Deserialize, ToSchema)]
#[serde(rename_all = "snake_case")]
pub enum BytedanceSeedreamV4MultiFunctionImageGenMaxImages {
One, // Default
Two,
Three,
Four,
}
#[derive(Copy, Clone, Debug, Serialize, Deserialize, ToSchema)]
#[serde(rename_all = "snake_case")]
pub enum BytedanceSeedreamV4MultiFunctionImageGenImageSize {
// Square
Square,
SquareHd,
// Tall
PortraitFourThree,
PortraitSixteenNine,
// Wide
LandscapeFourThree,
LandscapeSixteenNine,
// Auto
Auto,
Auto2k,
Auto4k,
}
#[derive(Serialize, Deserialize, ToSchema)]
pub struct BytedanceSeedreamV4MultiFunctionImageGenResponse {
pub success: bool,
pub inference_job_token: InferenceJobToken,
}
@@ -1,68 +0,0 @@
use serde_derive::{Deserialize, Serialize};
use tokens::tokens::generic_inference_jobs::InferenceJobToken;
use tokens::tokens::media_files::MediaFileToken;
use utoipa::ToSchema;
pub const BYTEDANCE_SEEDREAM_V4P5_MULTI_FUNCTION_IMAGE_GEN_PATH: &str = "/v1/generate/image/multi_function/bytedance_seedream_4p5";
#[derive(Serialize, Deserialize, ToSchema)]
pub struct BytedanceSeedreamV4p5MultiFunctionImageGenRequest {
/// Idempotency token to prevent duplicate requests.
pub uuid_idempotency_token: String,
/// Text prompt to generate the image from.
pub prompt: Option<String>,
/// Image media tokens to include in the editing context.
/// If present, we're doing image editing (image-to-image / image-editing)
/// If absent, we're doing image generation (text-to-image)
pub image_media_tokens: Option<Vec<MediaFileToken>>,
/// Number of images to generate. Default is one.
pub num_images: Option<BytedanceSeedreamV4p5MultiFunctionImageGenNumImages>,
pub max_images: Option<BytedanceSeedreamV4p5MultiFunctionImageGenMaxImages>,
pub image_size: Option<BytedanceSeedreamV4p5MultiFunctionImageGenImageSize>,
}
#[derive(Copy, Clone, Debug, Serialize, Deserialize, ToSchema)]
#[serde(rename_all = "snake_case")]
pub enum BytedanceSeedreamV4p5MultiFunctionImageGenNumImages {
One, // Default
Two,
Three,
Four,
}
#[derive(Copy, Clone, Debug, Serialize, Deserialize, ToSchema)]
#[serde(rename_all = "snake_case")]
pub enum BytedanceSeedreamV4p5MultiFunctionImageGenMaxImages {
One, // Default
Two,
Three,
Four,
}
#[derive(Copy, Clone, Debug, Serialize, Deserialize, ToSchema)]
#[serde(rename_all = "snake_case")]
pub enum BytedanceSeedreamV4p5MultiFunctionImageGenImageSize {
// Square
Square,
SquareHd,
// Tall
PortraitFourThree,
PortraitSixteenNine,
// Wide
LandscapeFourThree,
LandscapeSixteenNine,
// Auto
Auto2k,
Auto4k,
}
#[derive(Serialize, Deserialize, ToSchema)]
pub struct BytedanceSeedreamV4p5MultiFunctionImageGenResponse {
pub success: bool,
pub inference_job_token: InferenceJobToken,
}
@@ -1,95 +0,0 @@
use serde_derive::{Deserialize, Serialize};
use tokens::tokens::generic_inference_jobs::InferenceJobToken;
use tokens::tokens::media_files::MediaFileToken;
use utoipa::ToSchema;
pub const GPT_IMAGE_1P5_MULTI_FUNCTION_IMAGE_GEN_PATH: &str = "/v1/generate/image/multi_function/gpt_image_1p5";
#[derive(Serialize, Deserialize, ToSchema)]
pub struct GptImage1p5MultiFunctionImageGenRequest {
/// Idempotency token to prevent duplicate requests.
pub uuid_idempotency_token: String,
/// Text prompt to generate the image from.
pub prompt: Option<String>,
/// Image media tokens to include in the editing context (**OR INPAINTING CONTEXT!**).
/// If present, we're doing image editing (image-to-image / image-editing / image-inpainting)
/// If absent, we're doing image generation (text-to-image)
pub image_media_tokens: Option<Vec<MediaFileToken>>,
/// Only for image editing, which turns this into an inpainting task.
/// Text to image should not set this.
pub mask_image_token: Option<MediaFileToken>,
/// Number of images to generate. Default is one.
pub num_images: Option<GptImage1p5MultiFunctionImageGenNumImages>,
pub image_size: Option<GptImage1p5MultiFunctionImageGenSize>,
pub background: Option<GptImage1p5MultiFunctionImageGenBackground>,
pub quality: Option<GptImage1p5MultiFunctionImageGenQuality>,
/// Only for image editing.
pub input_fidelity: Option<GptImage1p5MultiFunctionImageGenInputFidelity>,
pub output_format: Option<GptImage1p5MultiFunctionImageGenOutputFormat>,
}
#[derive(Copy, Clone, Debug, Serialize, Deserialize, ToSchema)]
#[serde(rename_all = "snake_case")]
pub enum GptImage1p5MultiFunctionImageGenNumImages {
One, // Default
Two,
Three,
Four,
}
#[derive(Copy, Clone, Debug, Serialize, Deserialize, ToSchema)]
#[serde(rename_all = "snake_case")]
pub enum GptImage1p5MultiFunctionImageGenSize {
/// 1024x1024
Square,
/// 1536x1024
Wide,
/// 1024x1536
Tall,
}
#[derive(Copy, Clone, Debug, Serialize, Deserialize, ToSchema)]
#[serde(rename_all = "snake_case")]
pub enum GptImage1p5MultiFunctionImageGenBackground {
Auto,
Transparent,
Opaque,
}
#[derive(Copy, Clone, Debug, Serialize, Deserialize, ToSchema)]
#[serde(rename_all = "snake_case")]
pub enum GptImage1p5MultiFunctionImageGenQuality {
Low,
Medium,
High,
}
#[derive(Copy, Clone, Debug, Serialize, Deserialize, ToSchema)]
#[serde(rename_all = "snake_case")]
pub enum GptImage1p5MultiFunctionImageGenInputFidelity {
Low,
High,
}
#[derive(Copy, Clone, Debug, Serialize, Deserialize, ToSchema)]
#[serde(rename_all = "snake_case")]
pub enum GptImage1p5MultiFunctionImageGenOutputFormat {
Jpeg,
Png,
Webp,
}
#[derive(Serialize, Deserialize, ToSchema)]
pub struct GptImage1p5MultiFunctionImageGenResponse {
pub success: bool,
pub inference_job_token: InferenceJobToken,
}
@@ -1,5 +0,0 @@
pub mod bytedance_seedream_v4_multi_function_image_gen;
pub mod bytedance_seedream_v4p5_multi_function_image_gen;
pub mod gpt_image_1p5_multi_function_image_gen;
pub mod nano_banana_multi_function_image_gen;
pub mod nano_banana_pro_multi_function_image_gen;
@@ -1,62 +0,0 @@
use serde_derive::{Deserialize, Serialize};
use tokens::tokens::generic_inference_jobs::InferenceJobToken;
use tokens::tokens::media_files::MediaFileToken;
use utoipa::ToSchema;
pub const NANO_BANANA_MULTI_FUNCTION_IMAGE_GEN_PATH: &str = "/v1/generate/image/multi_function/nano_banana";
#[derive(Serialize, Deserialize, ToSchema, Debug)]
pub struct NanoBananaMultiFunctionImageGenRequest {
/// Idempotency token to prevent duplicate requests.
pub uuid_idempotency_token: String,
/// Text prompt to generate the image from.
pub prompt: Option<String>,
/// Image media tokens to include in the editing context.
/// If present, we're doing image editing (image-to-image / image-editing)
/// If absent, we're doing image generation (text-to-image)
pub image_media_tokens: Option<Vec<MediaFileToken>>,
/// Number of images to generate. Default is one.
pub num_images: Option<NanoBananaMultiFunctionImageGenNumImages>,
/// Aspect ratio of the images to generate. Default is "1:1"
pub aspect_ratio: Option<NanoBananaMultiFunctionImageGenAspectRatio>,
}
#[derive(Copy, Clone, Debug, Serialize, Deserialize, ToSchema)]
#[serde(rename_all = "snake_case")]
pub enum NanoBananaMultiFunctionImageGenNumImages {
One, // Default
Two,
Three,
Four,
}
#[derive(Copy, Clone, Debug, Serialize, Deserialize, ToSchema)]
#[serde(rename_all = "snake_case")]
pub enum NanoBananaMultiFunctionImageGenAspectRatio {
// Auto (only for image editing)
Auto, // Default for image editing.
// Square
OneByOne, // Default for text-to-image
// Wide
FiveByFour,
FourByThree,
ThreeByTwo,
SixteenByNine,
TwentyOneByNine,
// Tall
FourByFive,
ThreeByFour,
TwoByThree,
NineBySixteen, // NB: No NineByTwentyOne ?
}
#[derive(Serialize, Deserialize, ToSchema)]
pub struct NanoBananaMultiFunctionImageGenResponse {
pub success: bool,
pub inference_job_token: InferenceJobToken,
}
@@ -1,73 +0,0 @@
use serde_derive::{Deserialize, Serialize};
use tokens::tokens::generic_inference_jobs::InferenceJobToken;
use tokens::tokens::media_files::MediaFileToken;
use utoipa::ToSchema;
pub const NANO_BANANA_PRO_MULTI_FUNCTION_IMAGE_GEN_PATH: &str = "/v1/generate/image/multi_function/nano_banana_pro";
#[derive(Serialize, Deserialize, ToSchema, Debug)]
pub struct NanoBananaProMultiFunctionImageGenRequest {
/// Idempotency token to prevent duplicate requests.
pub uuid_idempotency_token: String,
/// Text prompt to generate the image from.
pub prompt: Option<String>,
/// Image media tokens to include in the editing context.
/// If present, we're doing image editing (image-to-image / image-editing)
/// If absent, we're doing image generation (text-to-image)
pub image_media_tokens: Option<Vec<MediaFileToken>>,
/// Number of images to generate. Default is one.
pub num_images: Option<NanoBananaProMultiFunctionImageGenNumImages>,
/// Resolution of the image to generate. Default is OneK (1K).
pub resolution: Option<NanoBananaProMultiFunctionImageGenImageResolution>,
/// Aspect ratio of the images to generate. Default is "1:1"
pub aspect_ratio: Option<NanoBananaProMultiFunctionImageGenAspectRatio>,
}
#[derive(Copy, Clone, Debug, Serialize, Deserialize, ToSchema)]
#[serde(rename_all = "snake_case")]
pub enum NanoBananaProMultiFunctionImageGenNumImages {
One, // Default
Two,
Three,
Four,
}
#[derive(Copy, Clone, Debug, Serialize, Deserialize, ToSchema)]
#[serde(rename_all = "snake_case")]
pub enum NanoBananaProMultiFunctionImageGenImageResolution {
OneK,
TwoK,
FourK,
}
#[derive(Copy, Clone, Debug, Serialize, Deserialize, ToSchema)]
#[serde(rename_all = "snake_case")]
pub enum NanoBananaProMultiFunctionImageGenAspectRatio {
// Auto (only for image editing)
Auto, // Default for image editing.
// Square
OneByOne, // Default for text-to-image
// Wide
FiveByFour,
FourByThree,
ThreeByTwo,
SixteenByNine,
TwentyOneByNine,
// Tall
FourByFive,
ThreeByFour,
TwoByThree,
NineBySixteen, // NB: No NineByTwentyOne ?
}
#[derive(Serialize, Deserialize, ToSchema)]
pub struct NanoBananaProMultiFunctionImageGenResponse {
pub success: bool,
pub inference_job_token: InferenceJobToken,
}
@@ -1,50 +0,0 @@
use serde_derive::{Deserialize, Serialize};
use tokens::tokens::generic_inference_jobs::InferenceJobToken;
use tokens::tokens::media_files::MediaFileToken;
use utoipa::ToSchema;
pub const GEMINI_25_FLASH_EDIT_IMAGE_PATH: &str = "/v1/generate/image/gemini_25_flash";
#[derive(Serialize, Deserialize, ToSchema)]
pub struct Gemini25FlashTextToImageImageRequest {
/// Idempotency token to prevent duplicate requests.
pub uuid_idempotency_token: String,
/// Text prompt to generate the image from.
pub prompt: Option<String>,
// TODO: OneOf for image_urls or image_media_tokens
/// Image media tokens to include in the editing context.
pub image_media_tokens: Option<Vec<MediaFileToken>>,
/// Number of images to generate. Default is one.
pub num_images: Option<Gemini25FlashTextToImageImageNumImages>,
/// Quality of the image to generate. Default is High.
pub image_quality: Option<Gemini25FlashTextToImageImageImageQuality>,
}
#[derive(Copy, Clone, Debug, Serialize, Deserialize, ToSchema)]
#[serde(rename_all = "snake_case")]
pub enum Gemini25FlashTextToImageImageNumImages {
One, // Default
Two,
Three,
Four,
}
#[derive(Copy, Clone, Debug, Serialize, Deserialize, ToSchema)]
#[serde(rename_all = "snake_case")]
pub enum Gemini25FlashTextToImageImageImageQuality {
Auto,
Low,
Medium,
High, // Default
}
#[derive(Serialize, Deserialize, ToSchema)]
pub struct Gemini25FlashTextToImageImageResponse {
pub success: bool,
pub inference_job_token: InferenceJobToken,
}
@@ -1,3 +0,0 @@
pub mod generate_hunyuan_2_0_image_to_3d;
pub mod generate_hunyuan_2_1_image_to_3d;
pub mod multi_function;
@@ -1,84 +0,0 @@
use serde_derive::{Deserialize, Serialize};
use tokens::tokens::generic_inference_jobs::InferenceJobToken;
use tokens::tokens::media_files::MediaFileToken;
use utoipa::ToSchema;
pub const HUNYUAN3D_V3_MULTI_FUNCTION_OBJECT_GEN_PATH: &str = "/v1/generate/object/multi_function/hunyuan3d_v3";
/// Hunyuan 3D v3 Multi-Function Request
///
/// This endpoint combines three modes:
/// - Text-to-3D: Provide only a prompt (no image_media_token)
/// - Image-to-3D: Provide an image_media_token (no prompt or empty prompt)
/// - Sketch-to-3D: Provide both a prompt AND an image_media_token
#[derive(Serialize, Deserialize, ToSchema, Debug)]
pub struct Hunyuan3dV3MultiFunctionObjectGenRequest {
/// Idempotency token to prevent duplicate requests.
pub uuid_idempotency_token: String,
/// Text prompt for generation.
/// - Required for text-to-3d mode
/// - Required for sketch-to-3d mode
/// - Ignored for image-to-3d mode
pub prompt: Option<String>,
/// Primary image media token.
/// - Required for image-to-3d mode
/// - Required for sketch-to-3d mode
/// - Not used for text-to-3d mode
pub image_media_token: Option<MediaFileToken>,
/// Back view image for multi-view image-to-3d mode.
/// Only applicable when image_media_token is provided without a prompt.
pub back_image_media_token: Option<MediaFileToken>,
/// Left view image for multi-view image-to-3d mode.
/// Only applicable when image_media_token is provided without a prompt.
pub left_image_media_token: Option<MediaFileToken>,
/// Right view image for multi-view image-to-3d mode.
/// Only applicable when image_media_token is provided without a prompt.
pub right_image_media_token: Option<MediaFileToken>,
/// Target face count for the generated mesh.
/// Optional - if not provided, the model will use its default.
pub face_count: Option<u32>,
/// Type of generation to perform.
/// Default is Normal.
pub generate_type: Option<Hunyuan3dV3GenerateType>,
/// Type of polygon to use in the generated mesh.
/// Default is Triangle.
pub polygon_type: Option<Hunyuan3dV3PolygonType>,
/// Whether to enable PBR (Physically Based Rendering) materials.
/// Default is false.
pub enable_pbr: Option<bool>,
}
#[derive(Copy, Clone, Debug, Serialize, Deserialize, ToSchema)]
#[serde(rename_all = "snake_case")]
pub enum Hunyuan3dV3GenerateType {
/// Standard generation quality (default)
Normal,
/// Low polygon count for game assets
LowPoly,
/// Geometry only, no textures
Geometry,
}
#[derive(Copy, Clone, Debug, Serialize, Deserialize, ToSchema)]
#[serde(rename_all = "snake_case")]
pub enum Hunyuan3dV3PolygonType {
/// Triangle polygons (default)
Triangle,
/// Quadrilateral polygons
Quadrilateral,
}
#[derive(Serialize, Deserialize, ToSchema)]
pub struct Hunyuan3dV3MultiFunctionObjectGenResponse {
pub success: bool,
pub inference_job_token: InferenceJobToken,
}
@@ -1 +0,0 @@
pub mod hunyuan3d_v3_multi_function_object_gen;
@@ -1,64 +0,0 @@
use serde::{Deserialize, Serialize};
use tokens::tokens::generic_inference_jobs::InferenceJobToken;
use tokens::tokens::media_files::MediaFileToken;
use utoipa::ToSchema;
pub const KLING_2P5_TURBO_PRO_MULTI_FUNCTION_VIDEO_VIDEO_PATH: &str = "/v1/generate/video/multi_function/kling_2p5_turbo_pro";
/// Both text-to-video and image-to-video in one request.
#[derive(Serialize, Deserialize, ToSchema)]
pub struct Kling2p5TurboProMultiFunctionVideoGenRequest {
/// Idempotency token to prevent duplicate requests.
pub uuid_idempotency_token: String,
/// "Required".
/// Required downstream, but we'll coerce null to empty string.
/// Text prompt.
pub prompt: Option<String>,
/// Optional.
/// Only for text-to-video.
pub negative_prompt: Option<String>,
/// Optional.
/// Only for image-to-video
/// Source of the starting frame
/// If present, we're doing image-to-video
/// If absent, we're doing text-to-video
pub start_frame_image_media_token: Option<MediaFileToken>,
/// Optional.
/// Only for image-to-video
/// Source of the ending frame (if desired)
pub end_frame_image_media_token: Option<MediaFileToken>,
/// Optional
/// Duration of the video
pub duration: Option<Kling2p5TurboProMultiFunctionVideoGenDuration>,
/// Optional.
/// Only for text-to-video
pub aspect_ratio: Option<Kling2p5TurboProMultiFunctionVideoGenAspectRatio>,
}
#[derive(Serialize, Deserialize, ToSchema)]
#[serde(rename_all = "snake_case")]
pub enum Kling2p5TurboProMultiFunctionVideoGenDuration {
FiveSeconds,
TenSeconds,
}
#[derive(Serialize, Deserialize, ToSchema)]
#[serde(rename_all = "snake_case")]
pub enum Kling2p5TurboProMultiFunctionVideoGenAspectRatio {
Square,
SixteenByNine,
NineBySixteen,
}
#[derive(Serialize, Deserialize, ToSchema)]
pub struct Kling2p5TurboProMultiFunctionVideoGenResponse {
pub success: bool,
pub inference_job_token: InferenceJobToken,
}
@@ -1,68 +0,0 @@
use serde::{Deserialize, Serialize};
use tokens::tokens::generic_inference_jobs::InferenceJobToken;
use tokens::tokens::media_files::MediaFileToken;
use utoipa::ToSchema;
pub const KLING_2P6_PRO_MULTI_FUNCTION_VIDEO_VIDEO_PATH: &str = "/v1/generate/video/multi_function/kling_2p6_pro";
/// Both text-to-video and image-to-video in one request.
#[derive(Serialize, Deserialize, ToSchema)]
pub struct Kling2p6ProMultiFunctionVideoGenRequest {
/// Idempotency token to prevent duplicate requests.
pub uuid_idempotency_token: String,
/// "Required".
/// Required downstream, but we'll coerce null to empty string.
/// Text prompt.
pub prompt: Option<String>,
/// Optional.
/// Only for text-to-video.
pub negative_prompt: Option<String>,
/// Optional.
/// Only for image-to-video
/// Source of the starting frame
/// If present, we're doing image-to-video
/// If absent, we're doing text-to-video
pub start_frame_image_media_token: Option<MediaFileToken>,
// /// Optional.
// /// Only for image-to-video
// /// Source of the ending frame (if desired)
// pub end_frame_image_media_token: Option<MediaFileToken>,
/// Optional
/// Duration of the video
pub duration: Option<Kling2p6ProMultiFunctionVideoGenDuration>,
/// Optional
/// Whether to generate audio.
pub generate_audio: Option<bool>,
/// Optional.
/// Only for text-to-video
pub aspect_ratio: Option<Kling2p6ProMultiFunctionVideoGenAspectRatio>,
}
#[derive(Serialize, Deserialize, ToSchema)]
#[serde(rename_all = "snake_case")]
pub enum Kling2p6ProMultiFunctionVideoGenDuration {
FiveSeconds,
TenSeconds,
}
#[derive(Serialize, Deserialize, ToSchema)]
#[serde(rename_all = "snake_case")]
pub enum Kling2p6ProMultiFunctionVideoGenAspectRatio {
Square,
SixteenByNine,
NineBySixteen,
}
#[derive(Serialize, Deserialize, ToSchema)]
pub struct Kling2p6ProMultiFunctionVideoGenResponse {
pub success: bool,
pub inference_job_token: InferenceJobToken,
}
@@ -1,6 +0,0 @@
pub mod kling_2_5_turbo_multi_function_video_gen;
pub mod kling_2_6_multi_function_video_gen;
pub mod sora_2_multi_function_video_gen;
pub mod sora_2_pro_multi_function_video_gen;
pub mod veo_3p1_fast_multi_function_video_gen;
pub mod veo_3p1_multi_function_video_gen;
@@ -1,66 +0,0 @@
use serde::{Deserialize, Serialize};
use tokens::tokens::generic_inference_jobs::InferenceJobToken;
use tokens::tokens::media_files::MediaFileToken;
use utoipa::ToSchema;
pub const SORA_2_MULTI_FUNCTION_VIDEO_VIDEO_PATH: &str = "/v1/generate/video/multi_function/sora_2";
/// Both text-to-video and image-to-video in one request.
#[derive(Serialize, Deserialize, ToSchema)]
pub struct Sora2MultiFunctionVideoGenRequest {
/// Idempotency token to prevent duplicate requests.
pub uuid_idempotency_token: String,
/// "Required".
/// Required downstream, but we'll coerce null to empty string.
/// Text prompt.
pub prompt: Option<String>,
/// Optional.
/// Only for image-to-video
/// Source of the starting frame
/// If present, we're doing image-to-video
/// If absent, we're doing text-to-video
pub image_media_token: Option<MediaFileToken>,
/// Optional.
pub resolution: Option<Sora2MultiFunctionVideoGenResolution>,
/// Optional
/// Duration of the video
/// (this is uniform against all modes)
pub duration: Option<Sora2MultiFunctionVideoGenDuration>,
/// Optional.
pub aspect_ratio: Option<Sora2MultiFunctionVideoGenAspectRatio>,
}
#[derive(Serialize, Deserialize, ToSchema)]
#[serde(rename_all = "snake_case")]
pub enum Sora2MultiFunctionVideoGenResolution {
Auto,
SevenTwentyP,
}
#[derive(Serialize, Deserialize, ToSchema)]
#[serde(rename_all = "snake_case")]
pub enum Sora2MultiFunctionVideoGenDuration {
FourSeconds,
EightSeconds,
TwelveSeconds,
}
#[derive(Serialize, Deserialize, ToSchema)]
#[serde(rename_all = "snake_case")]
pub enum Sora2MultiFunctionVideoGenAspectRatio {
Auto,
NineBySixteen,
SixteenByNine,
}
#[derive(Serialize, Deserialize, ToSchema)]
pub struct Sora2MultiFunctionVideoGenResponse {
pub success: bool,
pub inference_job_token: InferenceJobToken,
}
@@ -1,67 +0,0 @@
use serde::{Deserialize, Serialize};
use tokens::tokens::generic_inference_jobs::InferenceJobToken;
use tokens::tokens::media_files::MediaFileToken;
use utoipa::ToSchema;
pub const SORA_2_PRO_MULTI_FUNCTION_VIDEO_VIDEO_PATH: &str = "/v1/generate/video/multi_function/sora_2_pro";
/// Both text-to-video and image-to-video in one request.
#[derive(Serialize, Deserialize, ToSchema)]
pub struct Sora2ProMultiFunctionVideoGenRequest {
/// Idempotency token to prevent duplicate requests.
pub uuid_idempotency_token: String,
/// "Required".
/// Required downstream, but we'll coerce null to empty string.
/// Text prompt.
pub prompt: Option<String>,
/// Optional.
/// Only for image-to-video
/// Source of the starting frame
/// If present, we're doing image-to-video
/// If absent, we're doing text-to-video
pub image_media_token: Option<MediaFileToken>,
/// Optional.
pub resolution: Option<Sora2ProMultiFunctionVideoGenResolution>,
/// Optional
/// Duration of the video
/// (this is uniform against all modes)
pub duration: Option<Sora2ProMultiFunctionVideoGenDuration>,
/// Optional.
pub aspect_ratio: Option<Sora2ProMultiFunctionVideoGenAspectRatio>,
}
#[derive(Serialize, Deserialize, ToSchema)]
#[serde(rename_all = "snake_case")]
pub enum Sora2ProMultiFunctionVideoGenResolution {
Auto,
SevenTwentyP,
TenEightyP,
}
#[derive(Serialize, Deserialize, ToSchema)]
#[serde(rename_all = "snake_case")]
pub enum Sora2ProMultiFunctionVideoGenDuration {
FourSeconds,
EightSeconds,
TwelveSeconds,
}
#[derive(Serialize, Deserialize, ToSchema)]
#[serde(rename_all = "snake_case")]
pub enum Sora2ProMultiFunctionVideoGenAspectRatio {
Auto,
NineBySixteen,
SixteenByNine,
}
#[derive(Serialize, Deserialize, ToSchema)]
pub struct Sora2ProMultiFunctionVideoGenResponse {
pub success: bool,
pub inference_job_token: InferenceJobToken,
}
@@ -1,90 +0,0 @@
use serde::{Deserialize, Serialize};
use tokens::tokens::generic_inference_jobs::InferenceJobToken;
use tokens::tokens::media_files::MediaFileToken;
use utoipa::ToSchema;
pub const VEO_3P1_FAST_MULTI_FUNCTION_VIDEO_VIDEO_PATH: &str = "/v1/generate/video/multi_function/veo_3p1_fast";
/// Both text-to-video and image-to-video in one request.
#[derive(Serialize, Deserialize, ToSchema)]
pub struct Veo3p1FastMultiFunctionVideoGenRequest {
/// Idempotency token to prevent duplicate requests.
pub uuid_idempotency_token: String,
/// "Required".
/// Required downstream, but we'll coerce null to empty string.
/// Text prompt.
pub prompt: Option<String>,
/// Optional.
/// Only for text-to-video.
pub negative_prompt: Option<String>,
/// Optional.
/// Only for image-to-video
/// Source of the starting frame
/// If present, we're doing image-to-video
/// If absent, we're doing text-to-video
pub start_frame_image_media_token: Option<MediaFileToken>,
/// Optional.
/// Only for image-to-video
/// Source of the ending frame (if desired)
pub end_frame_image_media_token: Option<MediaFileToken>,
/// Optional
/// Duration of the video
/// (this is uniform against all modes)
pub duration: Option<Veo3p1FastMultiFunctionVideoGenDuration>,
/// Optional.
pub aspect_ratio: Option<Veo3p1FastMultiFunctionVideoGenAspectRatio>,
/// Optional.
pub resolution: Option<Veo3p1FastMultiFunctionVideoGenResolution>,
/// Optional
/// Whether to generate audio.
pub generate_audio: Option<bool>,
/// Optional
/// Only for text-to-video
pub enhance_prompt: Option<bool>,
/// Optional
/// Only for text-to-video
pub seed: Option<i64>,
/// Optional
/// Only for text-to-video
pub auto_fix: Option<bool>,
}
#[derive(Serialize, Deserialize, ToSchema)]
#[serde(rename_all = "snake_case")]
pub enum Veo3p1FastMultiFunctionVideoGenDuration {
FourSeconds,
SixSeconds,
EightSeconds,
}
#[derive(Serialize, Deserialize, ToSchema)]
#[serde(rename_all = "snake_case")]
pub enum Veo3p1FastMultiFunctionVideoGenAspectRatio {
Auto,
SixteenByNine,
NineBySixteen,
}
#[derive(Serialize, Deserialize, ToSchema)]
#[serde(rename_all = "snake_case")]
pub enum Veo3p1FastMultiFunctionVideoGenResolution {
SevenTwentyP,
TenEightyP,
}
#[derive(Serialize, Deserialize, ToSchema)]
pub struct Veo3p1FastMultiFunctionVideoGenResponse {
pub success: bool,
pub inference_job_token: InferenceJobToken,
}
@@ -1,90 +0,0 @@
use serde::{Deserialize, Serialize};
use tokens::tokens::generic_inference_jobs::InferenceJobToken;
use tokens::tokens::media_files::MediaFileToken;
use utoipa::ToSchema;
pub const VEO_3P1_MULTI_FUNCTION_VIDEO_VIDEO_PATH: &str = "/v1/generate/video/multi_function/veo_3p1";
/// Both text-to-video and image-to-video in one request.
#[derive(Serialize, Deserialize, ToSchema)]
pub struct Veo3p1MultiFunctionVideoGenRequest {
/// Idempotency token to prevent duplicate requests.
pub uuid_idempotency_token: String,
/// "Required".
/// Required downstream, but we'll coerce null to empty string.
/// Text prompt.
pub prompt: Option<String>,
/// Optional.
/// Only for text-to-video.
pub negative_prompt: Option<String>,
/// Optional.
/// Only for image-to-video
/// Source of the starting frame
/// If present, we're doing image-to-video
/// If absent, we're doing text-to-video
pub start_frame_image_media_token: Option<MediaFileToken>,
/// Optional.
/// Only for image-to-video
/// Source of the ending frame (if desired)
pub end_frame_image_media_token: Option<MediaFileToken>,
/// Optional
/// Duration of the video
/// (this is uniform against all modes)
pub duration: Option<Veo3p1MultiFunctionVideoGenDuration>,
/// Optional.
pub aspect_ratio: Option<Veo3p1MultiFunctionVideoGenAspectRatio>,
/// Optional.
pub resolution: Option<Veo3p1MultiFunctionVideoGenResolution>,
/// Optional
/// Whether to generate audio.
pub generate_audio: Option<bool>,
/// Optional
/// Only for text-to-video
pub enhance_prompt: Option<bool>,
/// Optional
/// Only for text-to-video
pub seed: Option<i64>,
/// Optional
/// Only for text-to-video
pub auto_fix: Option<bool>,
}
#[derive(Serialize, Deserialize, ToSchema)]
#[serde(rename_all = "snake_case")]
pub enum Veo3p1MultiFunctionVideoGenDuration {
FourSeconds,
SixSeconds,
EightSeconds,
}
#[derive(Serialize, Deserialize, ToSchema)]
#[serde(rename_all = "snake_case")]
pub enum Veo3p1MultiFunctionVideoGenAspectRatio {
Auto,
SixteenByNine,
NineBySixteen,
}
#[derive(Serialize, Deserialize, ToSchema)]
#[serde(rename_all = "snake_case")]
pub enum Veo3p1MultiFunctionVideoGenResolution {
SevenTwentyP,
TenEightyP,
}
#[derive(Serialize, Deserialize, ToSchema)]
pub struct Veo3p1MultiFunctionVideoGenResponse {
pub success: bool,
pub inference_job_token: InferenceJobToken,
}
@@ -1,19 +0,0 @@
use serde_derive::{Deserialize, Serialize};
use utoipa::ToSchema;
use tokens::tokens::media_files::MediaFileToken;
#[derive(Serialize, Deserialize, ToSchema)]
pub struct DeleteMediaFileRequest {
pub set_delete: bool,
/// NB: this is only to disambiguate when a user is both a mod and an author.
pub as_mod: Option<bool>,
}
/// For the URL PathInfo
#[derive(Serialize, Deserialize, ToSchema)]
pub struct DeleteMediaFilePathInfo {
pub token: MediaFileToken,
}
@@ -1,12 +0,0 @@
use crate::common::responses::media_links::MediaLinks;
/// Get the thumbnail template for either images or videos.
pub fn media_links_to_thumbnail_template(links: &MediaLinks) -> Option<&str> {
// NB: We only populate video previews for video tasks.
if let Some(video) = links.maybe_video_previews.as_ref() {
return Some(&video.animated_thumbnail_template);
}
// Image case - use the image thumbnail template if it exists.
links.maybe_thumbnail_template.as_deref()
}
@@ -1 +0,0 @@
pub mod media_links_to_thumbnail_template;
@@ -1 +0,0 @@
pub mod remove_image_background;
@@ -1,5 +0,0 @@
pub mod bg_removal;
pub mod edit;
pub mod inpaint;
pub mod multi_function;
pub mod text;
@@ -1,18 +0,0 @@
use crate::credentials::storyteller_credential_set::StorytellerCredentialSet;
use crate::error::storyteller_error::StorytellerError;
use crate::utils::api_host::ApiHost;
use crate::utils::basic_json_post_request::basic_json_post_request;
use artcraft_api_defs::generate::image::multi_function::bytedance_seedream_v4_multi_function_image_gen::{BytedanceSeedreamV4MultiFunctionImageGenRequest, BytedanceSeedreamV4MultiFunctionImageGenResponse, BYTEDANCE_SEEDREAM_V4_MULTI_FUNCTION_IMAGE_GEN_PATH};
pub async fn bytedance_seedream_v4_multi_function_image_gen(
api_host: &ApiHost,
maybe_creds: Option<&StorytellerCredentialSet>,
request: BytedanceSeedreamV4MultiFunctionImageGenRequest,
) -> Result<BytedanceSeedreamV4MultiFunctionImageGenResponse, StorytellerError> {
Ok(basic_json_post_request(
api_host,
BYTEDANCE_SEEDREAM_V4_MULTI_FUNCTION_IMAGE_GEN_PATH,
maybe_creds,
request,
).await?)
}
@@ -1,18 +0,0 @@
use crate::credentials::storyteller_credential_set::StorytellerCredentialSet;
use crate::error::storyteller_error::StorytellerError;
use crate::utils::api_host::ApiHost;
use crate::utils::basic_json_post_request::basic_json_post_request;
use artcraft_api_defs::generate::image::multi_function::bytedance_seedream_v4p5_multi_function_image_gen::{BytedanceSeedreamV4p5MultiFunctionImageGenRequest, BytedanceSeedreamV4p5MultiFunctionImageGenResponse, BYTEDANCE_SEEDREAM_V4P5_MULTI_FUNCTION_IMAGE_GEN_PATH};
pub async fn bytedance_seedream_v4p5_multi_function_image_gen(
api_host: &ApiHost,
maybe_creds: Option<&StorytellerCredentialSet>,
request: BytedanceSeedreamV4p5MultiFunctionImageGenRequest,
) -> Result<BytedanceSeedreamV4p5MultiFunctionImageGenResponse, StorytellerError> {
Ok(basic_json_post_request(
api_host,
BYTEDANCE_SEEDREAM_V4P5_MULTI_FUNCTION_IMAGE_GEN_PATH,
maybe_creds,
request,
).await?)
}
@@ -1,18 +0,0 @@
use crate::credentials::storyteller_credential_set::StorytellerCredentialSet;
use crate::error::storyteller_error::StorytellerError;
use crate::utils::api_host::ApiHost;
use crate::utils::basic_json_post_request::basic_json_post_request;
use artcraft_api_defs::generate::image::multi_function::gpt_image_1p5_multi_function_image_gen::{GptImage1p5MultiFunctionImageGenRequest, GptImage1p5MultiFunctionImageGenResponse, GPT_IMAGE_1P5_MULTI_FUNCTION_IMAGE_GEN_PATH};
pub async fn gpt_image_1p5_multi_function_image_gen(
api_host: &ApiHost,
maybe_creds: Option<&StorytellerCredentialSet>,
request: GptImage1p5MultiFunctionImageGenRequest,
) -> Result<GptImage1p5MultiFunctionImageGenResponse, StorytellerError> {
Ok(basic_json_post_request(
api_host,
GPT_IMAGE_1P5_MULTI_FUNCTION_IMAGE_GEN_PATH,
maybe_creds,
request,
).await?)
}
@@ -1,5 +0,0 @@
pub mod bytedance_seedream_v4_multi_function_image_gen_image;
pub mod bytedance_seedream_v4p5_multi_function_image_gen_image;
pub mod gpt_image_1p5_multi_function_image_gen_image;
pub mod nano_banana_multi_function_image_gen_image;
pub mod nano_banana_pro_multi_function_image_gen_image;
@@ -1,18 +0,0 @@
use crate::credentials::storyteller_credential_set::StorytellerCredentialSet;
use crate::error::storyteller_error::StorytellerError;
use crate::utils::api_host::ApiHost;
use crate::utils::basic_json_post_request::basic_json_post_request;
use artcraft_api_defs::generate::image::multi_function::nano_banana_multi_function_image_gen::{NanoBananaMultiFunctionImageGenRequest, NanoBananaMultiFunctionImageGenResponse, NANO_BANANA_MULTI_FUNCTION_IMAGE_GEN_PATH};
pub async fn nano_banana_multi_function_image_gen(
api_host: &ApiHost,
maybe_creds: Option<&StorytellerCredentialSet>,
request: NanoBananaMultiFunctionImageGenRequest,
) -> Result<NanoBananaMultiFunctionImageGenResponse, StorytellerError> {
Ok(basic_json_post_request(
api_host,
NANO_BANANA_MULTI_FUNCTION_IMAGE_GEN_PATH,
maybe_creds,
request,
).await?)
}
@@ -1,18 +0,0 @@
use crate::credentials::storyteller_credential_set::StorytellerCredentialSet;
use crate::error::storyteller_error::StorytellerError;
use crate::utils::api_host::ApiHost;
use crate::utils::basic_json_post_request::basic_json_post_request;
use artcraft_api_defs::generate::image::multi_function::nano_banana_pro_multi_function_image_gen::{NanoBananaProMultiFunctionImageGenRequest, NanoBananaProMultiFunctionImageGenResponse, NANO_BANANA_PRO_MULTI_FUNCTION_IMAGE_GEN_PATH};
pub async fn nano_banana_pro_multi_function_image_gen(
api_host: &ApiHost,
maybe_creds: Option<&StorytellerCredentialSet>,
request: NanoBananaProMultiFunctionImageGenRequest,
) -> Result<NanoBananaProMultiFunctionImageGenResponse, StorytellerError> {
Ok(basic_json_post_request(
api_host,
NANO_BANANA_PRO_MULTI_FUNCTION_IMAGE_GEN_PATH,
maybe_creds,
request,
).await?)
}
@@ -1,3 +0,0 @@
pub mod generate_hunyuan_3d_2_0_image_to_3d;
pub mod generate_hunyuan_3d_2_1_image_to_3d;
pub mod multi_function;
@@ -1,18 +0,0 @@
use crate::credentials::storyteller_credential_set::StorytellerCredentialSet;
use crate::error::storyteller_error::StorytellerError;
use crate::utils::api_host::ApiHost;
use crate::utils::basic_json_post_request::basic_json_post_request;
use artcraft_api_defs::generate::object::multi_function::hunyuan3d_v3_multi_function_object_gen::{Hunyuan3dV3MultiFunctionObjectGenRequest, Hunyuan3dV3MultiFunctionObjectGenResponse, HUNYUAN3D_V3_MULTI_FUNCTION_OBJECT_GEN_PATH};
pub async fn hunyuan3d_v3_multi_function_object_gen(
api_host: &ApiHost,
maybe_creds: Option<&StorytellerCredentialSet>,
request: Hunyuan3dV3MultiFunctionObjectGenRequest,
) -> Result<Hunyuan3dV3MultiFunctionObjectGenResponse, StorytellerError> {
Ok(basic_json_post_request(
api_host,
HUNYUAN3D_V3_MULTI_FUNCTION_OBJECT_GEN_PATH,
maybe_creds,
request,
).await?)
}
@@ -1 +0,0 @@
pub mod hunyuan3d_v3_multi_function_object_gen;
@@ -1,18 +0,0 @@
use crate::credentials::storyteller_credential_set::StorytellerCredentialSet;
use crate::error::storyteller_error::StorytellerError;
use crate::utils::api_host::ApiHost;
use crate::utils::basic_json_post_request::basic_json_post_request;
use artcraft_api_defs::generate::video::multi_function::kling_2_5_turbo_multi_function_video_gen::{Kling2p5TurboProMultiFunctionVideoGenRequest, Kling2p5TurboProMultiFunctionVideoGenResponse, KLING_2P5_TURBO_PRO_MULTI_FUNCTION_VIDEO_VIDEO_PATH};
pub async fn kling_2p5_turbo_pro_multi_function_video_gen(
api_host: &ApiHost,
maybe_creds: Option<&StorytellerCredentialSet>,
request: Kling2p5TurboProMultiFunctionVideoGenRequest,
) -> Result<Kling2p5TurboProMultiFunctionVideoGenResponse, StorytellerError> {
Ok(basic_json_post_request(
api_host,
KLING_2P5_TURBO_PRO_MULTI_FUNCTION_VIDEO_VIDEO_PATH,
maybe_creds,
request,
).await?)
}
@@ -1,18 +0,0 @@
use crate::credentials::storyteller_credential_set::StorytellerCredentialSet;
use crate::error::storyteller_error::StorytellerError;
use crate::utils::api_host::ApiHost;
use crate::utils::basic_json_post_request::basic_json_post_request;
use artcraft_api_defs::generate::video::multi_function::kling_2_6_multi_function_video_gen::{Kling2p6ProMultiFunctionVideoGenRequest, Kling2p6ProMultiFunctionVideoGenResponse, KLING_2P6_PRO_MULTI_FUNCTION_VIDEO_VIDEO_PATH};
pub async fn kling_2p6_pro_multi_function_video_gen(
api_host: &ApiHost,
maybe_creds: Option<&StorytellerCredentialSet>,
request: Kling2p6ProMultiFunctionVideoGenRequest,
) -> Result<Kling2p6ProMultiFunctionVideoGenResponse, StorytellerError> {
Ok(basic_json_post_request(
api_host,
KLING_2P6_PRO_MULTI_FUNCTION_VIDEO_VIDEO_PATH,
maybe_creds,
request,
).await?)
}
@@ -1,6 +0,0 @@
pub mod kling_2p5_turbo_pro_multi_function_video_gen;
pub mod kling_2p6_pro_multi_function_video_gen;
pub mod sora_2_multi_function_video_gen;
pub mod sora_2_pro_multi_function_video_gen;
pub mod veo_3p1_fast_multi_function_video_gen;
pub mod veo_3p1_multi_function_video_gen;
@@ -1,18 +0,0 @@
use crate::credentials::storyteller_credential_set::StorytellerCredentialSet;
use crate::error::storyteller_error::StorytellerError;
use crate::utils::api_host::ApiHost;
use crate::utils::basic_json_post_request::basic_json_post_request;
use artcraft_api_defs::generate::video::multi_function::sora_2_multi_function_video_gen::{Sora2MultiFunctionVideoGenRequest, Sora2MultiFunctionVideoGenResponse, SORA_2_MULTI_FUNCTION_VIDEO_VIDEO_PATH};
pub async fn sora_2_multi_function_video_gen(
api_host: &ApiHost,
maybe_creds: Option<&StorytellerCredentialSet>,
request: Sora2MultiFunctionVideoGenRequest,
) -> Result<Sora2MultiFunctionVideoGenResponse, StorytellerError> {
Ok(basic_json_post_request(
api_host,
SORA_2_MULTI_FUNCTION_VIDEO_VIDEO_PATH,
maybe_creds,
request,
).await?)
}
@@ -1,18 +0,0 @@
use crate::credentials::storyteller_credential_set::StorytellerCredentialSet;
use crate::error::storyteller_error::StorytellerError;
use crate::utils::api_host::ApiHost;
use crate::utils::basic_json_post_request::basic_json_post_request;
use artcraft_api_defs::generate::video::multi_function::sora_2_pro_multi_function_video_gen::{Sora2ProMultiFunctionVideoGenRequest, Sora2ProMultiFunctionVideoGenResponse, SORA_2_PRO_MULTI_FUNCTION_VIDEO_VIDEO_PATH};
pub async fn sora_2_pro_multi_function_video_gen(
api_host: &ApiHost,
maybe_creds: Option<&StorytellerCredentialSet>,
request: Sora2ProMultiFunctionVideoGenRequest,
) -> Result<Sora2ProMultiFunctionVideoGenResponse, StorytellerError> {
Ok(basic_json_post_request(
api_host,
SORA_2_PRO_MULTI_FUNCTION_VIDEO_VIDEO_PATH,
maybe_creds,
request,
).await?)
}
@@ -1,18 +0,0 @@
use crate::credentials::storyteller_credential_set::StorytellerCredentialSet;
use crate::error::storyteller_error::StorytellerError;
use crate::utils::api_host::ApiHost;
use crate::utils::basic_json_post_request::basic_json_post_request;
use artcraft_api_defs::generate::video::multi_function::veo_3p1_fast_multi_function_video_gen::{Veo3p1FastMultiFunctionVideoGenRequest, Veo3p1FastMultiFunctionVideoGenResponse, VEO_3P1_FAST_MULTI_FUNCTION_VIDEO_VIDEO_PATH};
pub async fn veo_3p1_fast_multi_function_video_gen(
api_host: &ApiHost,
maybe_creds: Option<&StorytellerCredentialSet>,
request: Veo3p1FastMultiFunctionVideoGenRequest,
) -> Result<Veo3p1FastMultiFunctionVideoGenResponse, StorytellerError> {
Ok(basic_json_post_request(
api_host,
VEO_3P1_FAST_MULTI_FUNCTION_VIDEO_VIDEO_PATH,
maybe_creds,
request,
).await?)
}
@@ -1,18 +0,0 @@
use crate::credentials::storyteller_credential_set::StorytellerCredentialSet;
use crate::error::storyteller_error::StorytellerError;
use crate::utils::api_host::ApiHost;
use crate::utils::basic_json_post_request::basic_json_post_request;
use artcraft_api_defs::generate::video::multi_function::veo_3p1_multi_function_video_gen::{Veo3p1MultiFunctionVideoGenRequest, Veo3p1MultiFunctionVideoGenResponse, VEO_3P1_MULTI_FUNCTION_VIDEO_VIDEO_PATH};
pub async fn veo_3p1_multi_function_video_gen(
api_host: &ApiHost,
maybe_creds: Option<&StorytellerCredentialSet>,
request: Veo3p1MultiFunctionVideoGenRequest,
) -> Result<Veo3p1MultiFunctionVideoGenResponse, StorytellerError> {
Ok(basic_json_post_request(
api_host,
VEO_3P1_MULTI_FUNCTION_VIDEO_VIDEO_PATH,
maybe_creds,
request,
).await?)
}
@@ -1,37 +0,0 @@
use crate::credentials::storyteller_credential_set::StorytellerCredentialSet;
use crate::error::storyteller_error::StorytellerError;
use crate::utils::api_host::ApiHost;
use crate::utils::basic_json_delete_request::basic_json_delete_request;
use artcraft_api_defs::common::responses::simple_generic_json_success::SimpleGenericJsonSuccess;
use artcraft_api_defs::media_file::delete_media_file::DeleteMediaFileRequest;
use artcraft_api_defs::subscriptions::get_session_subscription::GetSessionSubscriptionResponse;
use enums::common::payments_namespace::PaymentsNamespace;
use log::debug;
use tokens::tokens::media_files::MediaFileToken;
pub async fn delete_media_file(
api_host: &ApiHost,
maybe_creds: Option<&StorytellerCredentialSet>,
media_file_token: &MediaFileToken,
) -> Result<SimpleGenericJsonSuccess, StorytellerError> {
let url_path = get_url_path(media_file_token);
debug!("Requesting {:?}", &url_path);
// NB: We probably don't need to expose these options from FakeYou anymore.
let request = DeleteMediaFileRequest {
set_delete: true,
as_mod: None,
};
Ok(basic_json_delete_request(
api_host,
&url_path,
maybe_creds,
request,
).await?)
}
fn get_url_path(token: &MediaFileToken) -> String {
format!("/v1/media_files/file/{}", token.as_str())
}
@@ -1,143 +0,0 @@
use crate::credentials::storyteller_credential_set::StorytellerCredentialSet;
use crate::error::api_error::ApiError;
use crate::error::client_error::ClientError;
use crate::error::storyteller_error::StorytellerError;
use crate::utils::api_host::ApiHost;
use crate::utils::constants::{APPLICATION_JSON, USER_AGENT};
use crate::utils::filter_bad_response::filter_bad_response;
use crate::utils::http_get_anonymous::http_get_anonymous;
use chrono::{DateTime, Utc};
use enums::by_table::media_files::media_file_animation_type::MediaFileAnimationType;
use enums::by_table::media_files::media_file_class::MediaFileClass;
use enums::by_table::media_files::media_file_engine_category::MediaFileEngineCategory;
use enums::by_table::media_files::media_file_type::MediaFileType;
use enums::by_table::model_weights::weights_category::WeightsCategory;
use enums::common::generation_provider::GenerationProvider;
use enums::common::visibility::Visibility;
use enums::no_table::style_transfer::style_transfer_name::StyleTransferName;
use enums_public::by_table::model_weights::public_weights_types::PublicWeightsType;
use idempotency::uuid::generate_random_uuid;
use log::debug;
use reqwest::multipart::Form;
use reqwest::Client;
use serde_derive::{Deserialize, Serialize};
use std::fs::File;
use std::io::Read;
use std::path::Path;
use tokens::tokens::batch_generations::BatchGenerationToken;
use tokens::tokens::media_files::MediaFileToken;
use tokens::tokens::model_weights::ModelWeightToken;
use tokens::tokens::prompts::PromptToken;
use uuid::uuid;
pub struct LegacyUploadMediaFileFromFileArgs<'a, P: AsRef<Path>> {
pub api_host: &'a ApiHost,
pub maybe_creds: Option<&'a StorytellerCredentialSet>,
// NB: Path needs to be owned for the request.
pub path: P,
// /// If true, we should hide the image from the user's gallery.
// pub is_intermediate_system_file: bool,
// /// If provided, this is the prompt that this image is associated with.
// /// NOTE: Cannot set `is_intermediate_system_file = true` if this is set.
// pub maybe_prompt_token: Option<&'a PromptToken>,
// /// If provided, this is the service provider that created the image.
// /// NOTE: Cannot set `is_intermediate_system_file = true` if this is set.
// pub maybe_generation_provider: Option<GenerationProvider>,
// /// If provided, this groups the file into a batch
// /// TODO: This shouldn't be set clientside without the backend generating the token
// /// and cryptographically securing it. But we need to go fast here.
// pub maybe_batch_token: Option<&'a BatchGenerationToken>,
}
/// Upload using the legacy endpoint (spz, etc.)
pub async fn legacy_upload_media_file_from_file<P: AsRef<Path>>(
args: LegacyUploadMediaFileFromFileArgs<'_, P>,
) -> Result<UploadImageMediaFileSuccessResponse, StorytellerError> {
validate_args(&args)?;
let url = get_route(args.api_host);
debug!("Requesting {:?}", &url);
//let mut file = File::open(path)?;
let client = Client::builder()
.gzip(true)
.build()
.map_err(|err| StorytellerError::Client(ClientError::from(err)))?;
let mut form = Form::new()
.text("uuid_idempotency_token", generate_random_uuid())
.file("file", args.path)
.await
.map_err(|err| StorytellerError::Client(ClientError::from(err)))?;
//if args.is_intermediate_system_file {
// form = form.text("is_intermediate_system_file", "true");
//}
//if let Some(prompt_token) = &args.maybe_prompt_token {
// form = form.text("maybe_prompt_token", prompt_token.to_string());
//}
//if let Some(batch_token) = &args.maybe_batch_token {
// form = form.text("maybe_batch_token", batch_token.to_string());
//}
let mut request_builder = client.post(url)
.header("User-Agent", USER_AGENT)
.header("Accept", APPLICATION_JSON);
if let Some(creds) = args.maybe_creds {
if let Some(header) = &creds.maybe_as_cookie_header() {
request_builder = request_builder.header("Cookie", header);
}
}
let response = request_builder
.multipart(form)
.send()
.await
.map_err(|err| StorytellerError::Api(ApiError::from(err)))?;
let response = filter_bad_response(response).await?;
let response_body = &response.text().await
.map_err(|err| StorytellerError::Api(ApiError::from(err)))?;
let media_file = serde_json::from_str(&response_body)
.map_err(|err| StorytellerError::Api(ApiError::from(err)))?;
Ok(media_file)
}
fn get_route(api_host: &ApiHost) -> String {
let api_hostname_and_scheme = api_host.to_api_hostname_and_scheme();
format!("{}/v1/media_files/upload", api_hostname_and_scheme)
}
// TODO(bt,2025-04-22): Share API definitions between client and server in common crate.
#[derive(Deserialize, Debug)]
pub struct UploadImageMediaFileSuccessResponse {
pub success: bool,
pub media_file_token: MediaFileToken,
}
fn validate_args<P: AsRef<Path>>(
args: &LegacyUploadMediaFileFromFileArgs<'_, P>,
) -> Result<(), StorytellerError> {
// if args.is_intermediate_system_file && args.maybe_prompt_token.is_some() {
// return Err(StorytellerError::Client(ClientError::InvalidPreflightRequest(
// "Cannot set `is_intermediate_system_file` to true if `maybe_prompt_token` is provided."
// .to_string())));
// }
Ok(())
}
@@ -1,70 +0,0 @@
use crate::credentials::storyteller_credential_set::StorytellerCredentialSet;
use crate::error::api_error::ApiError;
use crate::error::client_error::ClientError;
use crate::error::storyteller_error::StorytellerError;
use crate::utils::api_host::ApiHost;
use crate::utils::constants::{APPLICATION_JSON, USER_AGENT};
use crate::utils::filter_bad_response::filter_bad_response;
use log::{debug, info};
use reqwest::Client;
use serde::de::DeserializeOwned;
use serde::{Deserialize, Serialize};
pub async fn basic_json_delete_request<Req: Serialize, Res: DeserializeOwned>(
api_host: &ApiHost,
route_path: &str,
maybe_creds: Option<&StorytellerCredentialSet>,
request: Req,
) -> Result<Res, StorytellerError> {
let url = get_route(api_host, route_path);
info!("DELETE {:?}", &url);
let client = Client::builder()
.gzip(true)
.build()
.map_err(|err| StorytellerError::Client(ClientError::from(err)))?;
let mut request_builder = client.delete(url)
.header("User-Agent", USER_AGENT)
.header("Accept", APPLICATION_JSON)
.header("Content-Type", APPLICATION_JSON);
if let Some(creds) = maybe_creds {
if let Some(header) = &creds.maybe_as_cookie_header() {
request_builder = request_builder.header("Cookie", header);
}
}
let request_body = serde_json::to_string(&request)
.map_err(|err| StorytellerError::Client(ClientError::from(err)))?;
debug!("Request body: {:?}", request_body);
let response = request_builder
.body(request_body)
.send()
.await
.map_err(|err| StorytellerError::Api(ApiError::from(err)))?;
let response = filter_bad_response(response)
.await
.map_err(|err| StorytellerError::Api(ApiError::from(err)))?;
let response_body = &response.text()
.await
.map_err(|err| StorytellerError::Api(ApiError::from(err)))?;
debug!("Response body: {:?}", response_body);
let response = serde_json::from_str(&response_body)
.map_err(|err| StorytellerError::Api(ApiError::from(err)))?;
Ok(response)
}
fn get_route(api_host: &ApiHost, route_path: &str) -> String {
let api_hostname = api_host.to_api_hostname_and_scheme();
format!("{}{}", api_hostname, route_path)
}
@@ -1,9 +0,0 @@
pub (crate) mod basic_json_delete_request;
pub (crate) mod basic_json_get_request;
pub (crate) mod basic_json_post_request;
pub (crate) mod basic_query_string_post_request;
pub (crate) mod filter_bad_response;
pub (crate) mod http_get_anonymous;
pub mod api_host;
pub mod constants;
pub mod status_codes;
@@ -3,5 +3,4 @@ pub mod job_details;
pub mod media_file_cover_image_details;
pub mod media_links;
pub mod simple_entity_stats;
pub mod simple_generic_json_success;
pub mod user_details_light;
@@ -1,6 +1,6 @@
pub mod flux_pro_kontext_max_edit_image;
pub mod gemini_25_flash_edit_image;
pub mod gpt_image_1_edit_image;
pub mod gemini_25_flash_edit_image;
pub mod qwen_edit_image;
pub mod seededit_3_edit_image;
@@ -19,7 +19,6 @@ pub struct GenerateFluxPro11TextToImageRequest {
pub num_images: Option<GenerateFluxPro11TextToImageNumImages>,
}
// TODO(btw,2026-01-01): Fal.ai and Fal.rs client libraries seem to disagree about the options.
#[derive(Copy, Clone, Debug, Serialize, Deserialize, ToSchema)]
#[serde(rename_all = "snake_case")]
pub enum GenerateFluxPro11TextToImageAspectRatio {
@@ -23,14 +23,10 @@ pub struct GenerateFluxPro11UltraTextToImageRequest {
#[serde(rename_all = "snake_case")]
pub enum GenerateFluxPro11UltraTextToImageAspectRatio {
Square, // 1:1
LandscapeThreeByTwo, // 3:2
LandscapeFourByThree, // 4:3
LandscapeSixteenByNine, // 16:9
LandscapeTwentyOneByNine, // 21:9
PortraitTwoByThree, // 2:3
PortraitThreeByFour, // 3:4
PortraitNineBySixteen, // 9:16
PortraitNineByTwentyOne, // 9:21
//Custom { width: u32, height: u32 }, // TODO
}
@@ -1,6 +1,8 @@
pub mod gemini_25_flash_text_to_image;
pub mod edit;
pub mod generate_flux_1_dev_text_to_image;
pub mod generate_flux_1_schnell_text_to_image;
pub mod generate_flux_pro_11_text_to_image;
pub mod generate_flux_pro_11_ultra_text_to_image;
pub mod generate_gpt_image_1_text_to_image;
pub mod inpaint;
pub mod remove_image_background;
@@ -0,0 +1,2 @@
pub mod generate_hunyuan_2_0_image_to_3d;
pub mod generate_hunyuan_2_1_image_to_3d;

Some files were not shown because too many files have changed in this diff Show More