chore(readme): cleanup [skip ci]

Author: gadicc

CONTRIBUTING.md

@@ -2,12 +2,20 @@
 
 *Tips for development*
 
+1. [General Hints](#general)
 1. [Development / Editor Setup](#editors)
   1. [Visual Studio Code (vscode)](#vscode)
 1. [Using Buildkit](#buildkit)
 1. [Local HTTP(S) Caching Proxy](#caching)
 1. [Local S3 Server](#local-s3-server)
 
+<a name="general"></a>
+## General
+
+1. Run docker with `-it` to make it easier to stop container with `Ctrl-C`.
+1. If you get a `CUDA initialization: CUDA unknown error` after suspend,
+    just stop the container, `rmmod nvidia_uvm`, and restart.
+
 <a name="editors"></a>
 ## Editors
 

README.md

@@ -38,47 +38,36 @@ assumptions.  If anything is unclear, please open an issue.
 *[See the `dev` branch for the latest features.](https://github.com/kiri-art/docker-diffusers-api/tree/dev)
 **Pull Requests must be submitted against the dev branch.***
 
-## Usage:
+## Installation & Setup:
+
+Setup varies depending on your use case.
 
-Firstly, fork and clone this repo.
+1. **To run locally or on a *server*, with runtime downloads:**:
 
-Most of the configuration happens via docker build variables.  You can
-see all the options in the [Dockerfile](./Dockerfile), and edit them
-there directly, or set via docker command line or e.g. Banana's dashboard
-UI once support for build variables land (any day now).
+  `docker run --gpus all -p 8000:8000 -e HF_AUTH_TOKEN=$HF_AUTH_TOKEN gadicc/diffusers-api`.
 
-If you're only deploying one container, that's all you need!  If you
-intend to deploy multiple containers each with different variables
-(e.g. a few different models), you can edit the example
-[`scripts/permutations.yaml`](scripts/permutations.yaml)] file and
-run [`scripts/permute.sh`](scripts/permute.sh) to create a number
-of sub-repos in the `permutations` directory.
+  See the [guides for various cloud providers](https://forums.kiri.art/t/running-on-other-cloud-providers/89/7).
 
-Lastly, there's an option to set `MODEL_ID=ALL`, and *all* models will
-be downloaded, and switched at request time (great for dev, useless for
-serverless).
+1. **To run *serverless*, include the model at build time**.
 
-**Deploying to banana?** That's it!  You're done.  Commit your changes and push.
+  1. [docker-diffusers-api-build-download](https://github.com/kiri-art/docker-diffusers-api-build-download) (
+    [banana](https://forums.kiri.art/t/run-diffusers-api-on-banana-dev/103), others)
+  1. [docker-diffusers-api-runpod](https://github.com/kiri-art/docker-diffusers-api-runpod),
+  see the [guide](https://forums.kiri.art/t/run-diffusers-api-on-runpod-io/102)
 
-## Running locally / development:
+1. **Building from source**.
 
-**Building**
+  1. Fork / clone this repo.
+  1. `docker build -t gadicc/diffusers-api .`
+  1. See [CONTRIBUTING.md](./CONTRIBUTING.md) for more helpful hints.
 
-1. `docker build -t diffusers-api --build-arg HF_AUTH_TOKEN=$HF_AUTH_TOKEN .`
-1. See [CONTRIBUTING.md](./CONTRIBUTING.md) for more helpful hints.
-1. Note: your first build can take a really long time, depending on
-    your PC & network speed, and *especially when using the `CHECKPOINT_URL`
-    feature*.  Great time to grab a coffee or take a walk.
+*Other configurations are possible but these are the most common cases*
 
-**Running**
+Everything is set via docker build-args or environment variables.
 
-1. `docker run -it --gpus all -p 8000:8000 diffusers-api`
-1. Note: the `-it` is optional but makes it alot quicker/easier to stop the
-    container using `Ctrl-C`.
-1. If you get a `CUDA initialization: CUDA unknown error` after suspend,
-    just stop the container, `rmmod nvidia_uvm`, and restart.
+## Usage:
 
-## Sending requests
+See also [Testing](#testing) below.
 
 The container expects an `HTTP POST` request with the following JSON body:
 
@@ -93,6 +82,7 @@ The container expects an `HTTP POST` request with the following JSON body:
     "seed": 3239022079
   },
   "callInputs": {
+    // You can leave these out to use the default
     "MODEL_ID": "runwayml/stable-diffusion-v1-5",
     "PIPELINE": "StableDiffusionPipeline",
     "SCHEDULER": "LMSDiscreteScheduler",
@@ -101,19 +91,6 @@ The container expects an `HTTP POST` request with the following JSON body:
 }
 ```
 
-If you're using banana's SDK, it looks something like this:
-
-```js
-const out = await banana.run(apiKey, modelKey, { "modelInputs": modelInputs, "callInputs": callInputs });
-```
-
-NB: if you're coming from another banana starter repo, note that we
-explicitly name `modelInputs` above, and send a bigger object (with
-`modelInputs` and `callInputs` keys) for the banana-sdk's
-"modelInputs" argument.
-
-If provided, `init_image` and `mask_image` should be base64 encoded.
-
 **Schedulers**: docker-diffusers-api is simply a wrapper around diffusers,
 literally any scheduler included in diffusers will work out of the box,
 provided it can loaded with its default config and without requiring
@@ -123,6 +100,8 @@ schedulers are the most common and most well tested:
 `LMSDiscreteScheduler`, `DDIMScheduler`, `PNDMScheduler`,
 `EulerAncestralDiscreteScheduler`, `EulerDiscreteScheduler`.
 
+**Pipelines**:
+
 <a name="testing"></a>
 ## Examples and testing
 
@@ -155,42 +134,27 @@ Request took 3.0s (init: 2.4s, inference: 2.1s)
 The best example of course is https://kiri.art/ and it's
 [source code](https://github.com/kiri-art/stable-diffusion-react-nextjs-mui-pwa).
 
-
-
-## Troubleshooting
-
-* **403 Client Error: Forbidden for url**
-
-  Make sure you've accepted the license on the model card of the HuggingFace model
-  specified in `MODEL_ID`, and that you correctly passed `HF_AUTH_TOKEN` to the
-  container.
+## Help on [Official Forums](https://forums.kiri.art/c/docker-diffusers-api/16).
 
 ## Adding other Models
 
 You have two options.
 
-1. For a diffusers model, simply set the `MODEL_ID` docker build variable to the name
+1. For a diffusers model, simply set `MODEL_ID` build-var / call-arg to the name
   of the model hosted on HuggingFace, and it will be downloaded automatically at
   build time.
 
-1. For a non-diffusers model, simply set the `CHECKPOINT_URL` docker build variable
+1. For a non-diffusers model, simply set the `CHECKPOINT_URL` build-var / call-arg
   to the URL of a `.ckpt` file, which will be downloaded and converted to the diffusers
-  format automatically at build time.
-
-## Keeping forks up to date
-
-Per your personal preferences, rebase or merge, e.g.
+  format automatically at build time.  `CHECKPOINT_CONFIG_URL` can also be set.
 
-1. `git fetch upstream`
-1. `git merge upstream/main`
-1. `git push`
-
-Or, if you're confident, do it in one step with no confirmations:
+## Troubleshooting
 
-  `git fetch upstream && git merge upstream/main --no-edit && git push`
+* **403 Client Error: Forbidden for url**
 
-Check `scripts/permute.sh` and your git remotes, some URLs are hardcoded, I'll
-make this easier in a future release.
+  Make sure you've accepted the license on the model card of the HuggingFace model
+  specified in `MODEL_ID`, and that you correctly passed `HF_AUTH_TOKEN` to the
+  container.
 
 ## Event logs / performance data