2.0.2 | New package name

Package name is updated as deepinfra

2.0.1 | Improved performance

Features

• AutomaticSpeechRecognition, ImageClassification, Zero shot image classification and object detection performance has been improved.

2.0.0 | Base models

Features

• DeepInfra API key can be provided from the environment.
• Image classification models are available.
• Zero shot image classification models are available.

2.0.0-rc

Better developer experience is on the way!

1.6.2 | Fix in payload and responses in text-to-image models

What's Changed

• bugfix(text-to-image types): TextToImage fields differ from SDXL fields by @ichernev in #16
• Cog models + SDXL by @ichernev in #15

1.6.1 | Documentation

## Features

• Documentation is added to npm

1.6.0 | Text classification, fill mask and token classification

## Features

• 4 text classification models are implemented.
• 4 object detection model is implemented.
• 14 token classifications models are added.
• 14 fill mask models are implemented.

1.5.1 | License update

• License is changed to MIT.

1.5.0 | Automatic Speech Recognition

Automatic Speech Recognition models

New release featuring various ASR models:

• WhisperLarge: Large ASR model.
• WhisperBase: Basic ASR model.
• WhisperBaseEn: Basic English ASR model.
• WhisperMedium: Medium ASR model.
• WhisperMediumEn: Medium English ASR model.
• WhisperSmall: Small ASR model.
• WhisperSmallEn: Small English ASR model.
• WhisperTimestampedMedium: Timestamped medium ASR model.
• WhisperTimestampedMediumEn: Timestamped medium English ASR model.
• WhisperTiny: Very small ASR model.
• WhisperTinyEn: Very small English ASR model.
• Updated documentation and examples provided for usage.

1.4.3 | 6 text to image models

Description
SDXL is cool, but is not the whole. There are many available in deepinfra.

Features

• Deliberate
• DreamShaper
• OpenJourney
• StableDiffusion v1.4
• StableDiffusion v1.5
• StableDiffusion v2.1