Bash CLI for use with Replicate.com APIBash CLI for use with Replicate.com API
Program : replicate-cli by peter@forret.com
Version : v0.0.2 (2023-01-17 16:23)
Purpose : Bash CLI for use with Replicate.com API
Usage : replicate-cli [-h] [-q] [-v] [-f] [-l <log_dir>] [-t <tmp_dir>] [-C <COUNT>] [-G <GUIDANCE>] [-H <HEIGHT>] [-M <MODEL>] [-N <STEPS>] [-R <STRENGTH>] [-S <SCHEDULE>] [-T <REPLICATE_API_TOKEN>] [-V <VERSION>] [-W <WIDTH>
] [-X <SEED>] <action> <prompt?> <negative?>
Flags, options and parameters:
-h|--help : [flag] show usage [default: off]
-q|--quiet : [flag] no output [default: off]
-v|--verbose : [flag] also show debug messages [default: off]
-f|--force : [flag] do not ask for confirmation (always yes) [default: off]
-l|--log_dir <?> : [option] folder for log files [default: /home/pforret/log/replicate-cli]
-t|--tmp_dir <?> : [option] folder for temp files [default: .tmp]
-C|--COUNT <?> : [option] Number of outputs [default: 1]
-G|--GUIDANCE <?>: [option] Classifier-free guidance [default: 7.5]
-H|--HEIGHT <?> : [option] Image height [default: 512]
-M|--MODEL <?> : [option] Prediction model to use [default: stability-ai/stable-diffusion]
-N|--STEPS <?> : [option] Number of steps [default: 50]
-R|--STRENGTH <?>: [option] Prompt strength (0-1) [default: 0.8]
-S|--SCHEDULE <?>: [option] Scheduler [default: DPMSolverMultistep]
-T|--REPLICATE_API_TOKEN <?>: [option] Replicate.com API token
-V|--VERSION <?> : [option] Prediction model version
-W|--WIDTH <?> : [option] Image width [default: 512]
-X|--SEED <?> : [option] Force start seeding
<action> : [choice] action to perform [options: collections,models,predict,check,env,update]
<prompt> : [parameter] prompt (optional)
<negative> : [parameter] negative prompt (optional)
### TIPS & EXAMPLES
* use replicate-cli predict to ...
replicate-cli predict
* use replicate-cli collections to list all collections
replicate-cli collections
* use replicate-cli models to list all models for a collection
replicate-cli models
replicate-cli models image-to-text
replicate-cli models super-resolution
* use replicate-cli check to check if this script is ready to execute and what values the options/flags are
replicate-cli check
* use replicate-cli env to generate an example .env file
replicate-cli env > .env
* use replicate-cli update to update to the latest version
replicate-cli update
* >>> bash script created with pforret/bashew
> replicate-cli collections
✅ These are the collection names you can use for 'replicate-cli models [collection]'
diffusion-models : https://replicate.com/collections/diffusion-models
image-restoration : https://replicate.com/collections/image-restoration
image-to-text : https://replicate.com/collections/image-to-text
ml-makeovers : https://replicate.com/collections/ml-makeovers
style-transfer : https://replicate.com/collections/style-transfer
super-resolution : https://replicate.com/collections/super-resolution
text-to-image : https://replicate.com/collections/text-to-image
> replicate-cli models super-resolution
✅ These are the models you can use for 'replicate-cli -M [model] predict ...'
⏳ Collection: super-resolution
cjwbw/rudalle-sr : https://replicate.com/cjwbw/rudalle-sr
cjwbw/swin2sr : https://replicate.com/cjwbw/swin2sr
codeslake/refvsr-cvpr2022 : https://replicate.com/codeslake/refvsr-cvpr2022
jingyunliang/hcflow-sr : https://replicate.com/jingyunliang/hcflow-sr
jingyunliang/swinir : https://replicate.com/jingyunliang/swinir
longguangwang/arbsr : https://replicate.com/longguangwang/arbsr
nightmareai/latent-sr : https://replicate.com/nightmareai/latent-sr
nightmareai/real-esrgan : https://replicate.com/nightmareai/real-esrgan
raoumer/srrescgan : https://replicate.com/raoumer/srrescgan
wzx0826/lbnet : https://replicate.com/wzx0826/lbnet
xinntao/esrgan : https://replicate.com/xinntao/esrgan
xpixelgroup/hat : https://replicate.com/xpixelgroup/hat
with basher
$ basher install pforret/replicate-cli
or with git
$ git clone https://github.com/pforret/replicate-cli.git
$ cd replicate-cli
- Replicate API on replicate.com/docs/reference/http
- script created with bashew
© 2023 Peter Forret