Skip to content
@ai-forever

AI Forever

Creating ML for the future. AI projects you already know. We are non-profit organization with members from all over the world.

Popular repositories Loading

  1. Kandinsky-2 Kandinsky-2 Public

    Kandinsky 2 — multilingual text2image latent diffusion model

    Jupyter Notebook 2.8k 319

  2. ru-gpts ru-gpts Public

    Russian GPT3 models.

    Python 2.1k 434

  3. ru-dalle ru-dalle Public

    Generate images from texts. In Russian

    Jupyter Notebook 1.6k 241

  4. ghost ghost Public

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

    Python 1.6k 310

  5. Real-ESRGAN Real-ESRGAN Public

    PyTorch implementation of Real-ESRGAN model

    Python 656 223

  6. gigachain gigachain Public

    Forked from langchain-ai/langchain

    ⚡ Набор решений для разработки LLM-приложений на русском языке с поддержкой GigaChat ⚡

    Jupyter Notebook 590 87

Repositories

Showing 10 of 126 repositories
  • FRIDA-Decisions Public

    Structured decisions over Russian text in one encoder pass: choice, scale, yes/no, ranking

    ai-forever/FRIDA-Decisions's past year of commit activity
    Python 19 2 0 0 Updated Oct 6, 2026
  • gpt2giga Public

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

    ai-forever/gpt2giga's past year of commit activity
    Python 137 MIT 49 1 3 Updated Oct 6, 2026
  • ai-forever/langchain-gigachat's past year of commit activity
    Python 53 MIT 13 7 9 Updated Oct 6, 2026
  • gigachat Public

    GigaChat API access library

    ai-forever/gigachat's past year of commit activity
    Python 171 MIT 45 6 12 Updated Oct 6, 2026
  • ai-forever/harness-bench-fast's past year of commit activity
    Python 55 MIT 11 5 3 Updated Oct 6, 2026
  • RUMBA Public

    Russian User Memory Benchmark evaluation

    ai-forever/RUMBA's past year of commit activity
    Jupyter Notebook 1 MIT 0 0 0 Updated Oct 5, 2026
  • giga_agent Public

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

    ai-forever/giga_agent's past year of commit activity
    Python 206 MIT 45 17 16 Updated Oct 5, 2026
  • STARM Public Forked from sapientinc/HRM

    Recursive architecture for algorithmic tasks that repeatedly applies a single computational block. STARM outperforms models with larger parameter counts on Arithmetic, Conway's Game of Life, Maze, Sudoku, and ARC-AGI tasks

    ai-forever/STARM's past year of commit activity
    Python 136 Apache-2.0 1,839 0 0 Updated Oct 1, 2026
  • WebPageBench Public

    WebPageBench: Event-Level Verification and Controlled UI-Variant Generation for Web Agents

    ai-forever/WebPageBench's past year of commit activity
    Vue 1 Apache-2.0 0 0 0 Updated Sep 30, 2026
  • Zarya-Orthrus Public

    Training and evaluation of hybrid models with the Orthrus architecture

    ai-forever/Zarya-Orthrus's past year of commit activity
    Python 0 MIT 0 0 0 Updated Sep 23, 2026