Reza Sayar Claim this profile

4-bit AWQ INT4 + NVFP4 quantization toolkit for ByteDance Lance multimodal model (image + video)

Pure-Rust reimplementation of Vosk/Kaldi nnet3 chain ASR decoding (no libvosk/Kaldi/Python). Reproduces vosk exactly.

On-device streaming FastConformer (Persian, 6.9M) on ESP32-S3 N32R16V — the int4 [70,0] research spike

RL Environments for training persona-based compassionate Emotional Support Conversation agents

Central release hub for Shenava 1.0: open streaming Persian ASR, datasets, apps, benchmarks, and paper artifacts.

No description.

1,969-tool MCP server giving AI assistants full Godot 4 editor parity — create scenes, control running games, generate scripts, and navigate 1,969 tools progressively so even tiny local models don't get lost.

Drag-and-drop any MuJoCo MJCF model into Rerun, fully articulated. A reference data loader for rerun-io/rerun#8634.

https://shop.tetheria.ai/

Benchmark of AI agent platforms on the Reducto API + a standalone MCP server

No description.

An open-source, hand-worn variant of UMI for collecting bimanual manipulation data without a robot in the loop: modular for any parallel-jaw gripper.

No description.

Visualize, query, and stream to train on multimodal robotics data.

本项目为xiaozhi-esp32提供后端服务，帮助您快速搭建ESP32设备控制服务器。Backend service for xiaozhi-esp32, helps you quickly build an ESP32 device control server.

Hotword-boosted CTC prefix-beam decoder for on-device ASR — dependency-free pyctcdecode port (Rust)

Tiny, no-nonsense, self-contained, Tensorflow and ONNX inference

A repository for rerun

The agent that grows with you

Desktop Companion for Hermes Agent

Famous AI agent harness, now in with a native desktop app with GUI

Project N.O.M.A.D, is a self-contained, offline survival computer packed with critical tools, knowledge, and AI to keep you informed and empowered—anytime, anywhere.

UncGPT 2026 — NeurIPS Competition: six-track grounded caregiving benchmark on a shared persona-conditioned substrate. Companion to Hugging Face collection huggingface.co/collections/Reza2kn/uncgpt-2026-neurips-competition-6a081d4a2a5645643dc999b4

Free Robotics & 3D-Printing Community Program (Mérida) — public site + program tooling.

SGLang is a high-performance serving framework for large language models and multimodal models.

A text-to-speech (TTS), speech-to-text (STT) and speech-to-speech (STS) library built on Apple's MLX framework, providing efficient speech analysis on Apple Silicon.

No description.

MLX-VLM is a package for inference and fine-tuning of Vision Language Models (VLMs) on your Mac using MLX.

LeIsaac provides teleoperation functionality in IsaacLab using the SO101Leader (LeRobot), including data collection, data conversion, and subsequent policy training.

No description.

There can be more than Notion and Miro. AFFiNE(pronounced [ə‘fain]) is a next-gen knowledge base that brings planning, sorting and creating all together. Privacy first, open-source, customizable and ready to use.

freeCodeCamp.org's open-source codebase and curriculum. Learn to code for free.

Let's bake some (AI) stickers!

A course on aligning smol models.

No description.

No description.

Image pilot with the power of Real-Time Latent Consistency Model

Visual Ears: An open-source app dismantling the barriers of hearing and language. We're crafting a space where speech becomes sight, languages connect seamlessly, and everyone accesses inclusive communication.