@taesiri
taesiri
Follow
View taesiri's full-sized avatar
😎
Just Do it!
Mohammad Reza Taesiri taesiri
😎
Just Do it!
AGI ... one linear layer at a time
Highlights
- Pro
Starred repositories
Fully automatic censorship removal for language models
A community collection of OpenClaw use cases for making life easier.
Code for Paper: Benchmarking Multi-step Scientific Tool-use in LLM Agents
Use claude-code for free in the terminal, VSCode extension or via telegram like openclaw
General plug-and-play inference library for Recursive Language Models (RLMs), supporting various sandboxes.
Real-time global intelligence dashboard — AI-powered news aggregation, geopolitical monitoring, and infrastructure tracking in a unified situational awareness interface
Skills for the Gemini API, SDK and model/agent interactions
41
Updated Feb 13, 2026
An end-to-end open ecosystem for robot learning
Code2Worlds: Empowering Coding LLMs for 4D World Generation
We introduce VISTA-Bench, a systematic benchmark from multimodal perception, reasoning, to unimodal understanding domains. It evaluates visualized text understanding by contrasting pure-text and vi...
The agent-native LLM router powering OpenClaw — by BlockRun
Semantic Search & Call Graphs for AI Agents (100% Local)
Complete Claude Code configuration collection - agents, skills, hooks, commands, rules, MCPs. Battle-tested configs from an Anthropic hackathon winner.
Official Implementation of "CLI-Gym: Scalable CLI Task Generation via Agentic Environment Inversion"
Fast, Sharp & Reliable Agentic Intelligence
Code for "SceneSmith: Agentic Generation of Simulation-Ready Indoor Scenes"
Agent World Model: Infinity Synthetic Environments for Agentic Reinforcement Learning
VLA-JEPA: Enhancing Vision-Language-Action Model with Latent World Model
Harbor is a framework for running agent evaluations and creating and using RL environments.
A Python library for extracting structured information from unstructured text using LLMs with precise source grounding and interactive visualization.
Official code for "How2Everything: Mining the Web for How-To Procedures to Evaluate and Improve LLMs"
Coordinate agent teams in Claude Code using a Royal Navy inspired command structure and operating procedures. It provides structured sailing orders, battle plans, action stations, and a captain's l...