ai-forever

Kandinsky 2 — multilingual text2image latent diffusion model

Russian GPT3 models.

Generate images from texts. In Russian

A new one shot face swap approach for image and video domains

PyTorch implementation of Real-ESRGAN model

BERT-NER (nert-bert) with google bert https://github.com/google-research.

MoVQGAN - model for the image encoding and reconstruction

RUDOLPH: One Hyper-Tasking Transformer can be creative as DALL-E and GPT-3 and smart as CLIP

GigaAgent — это универсальный агент-оркестратор для решения широкого круга задач (ReAct + REPL)

Multilingual Generative Pretrained Model

KandinskyVideo — multilingual end-to-end text2video latent diffusion model

GigaChat API access library

SAGE: Spelling correction, corruption and evaluation for multiple languages

A new one shot head swapping approach

CLIP implementation for Russian language

Text and image to video generation: Kandinsky 4.0 (2024)

FastAPI proxy for OpenAI, Anthropic and Gemini compatible clients backed by the GigaChat API

Kandinsky x Deforum — generating short animations

Materials of the AI Journey 2020 competition dedicated to the recognition of Peter the Great's manuscripts, https://ai-journey.ru/contest/task01

Augmentex — a library for augmenting texts with errors

MERA (Multimodal Evaluation for Russian-language Architectures) is a new open benchmark for the Russian language for evaluating fundamental models.

Spring AI extension for GigaChat LLM

NLP model zoo for Russian