h-embodvis/turbovla
TurboVLA: Real-Time Vision-Language-Action Model at 32 Hz on an RTX 4090 with <1 GB VRAM
01 / MACHINE INTELLIGENCE
Models you can run. Agents you can inspect. Tools you can build on.
← All topicsTurboVLA: Real-Time Vision-Language-Action Model at 32 Hz on an RTX 4090 with <1 GB VRAM
qqr is an RL training framework for open-ended agents.
Deploys your OS, databases, and SSL on your VPS in just 10 minutes. Orchestrates a team of AI agents for coding, marketing, and sales. The built-in optimizer saves up to 90% on token costs, letting you build and manage your online business directly through chat. Fully open-source.
Visibility into AI agent activity on endpoints, with on-device detection, optional pre-action blocking, and forensic reconstruction.
14MB ACP-Rust Agent, Claude Code Plugin compatible, Dynamic Workflow, Goal, Artifacts, Free Web Search, full feature and better support!
Free multi-agent coding workbench: Grok/Codex/Claude/Cursor panes + TrueMemory (per-repo + global)
A collection of agent skills that help with various parts of building a great interface. From animation and UI polish to accessibility and product writing.
🐮 Build distributed, event-driven agents.
Deterministic, local-first memory and guardrails for AI coding agents with no LLM in the hot path.
Off Grid AI — private, on-device AI. Run open models (text, vision, image, voice) locally through one OpenAI-compatible gateway. No cloud, no accounts, no API keys.
Agent multiplexer with map-charting capabilities.
MCP server for creating pixel art with Aseprite through AI assistants. Supports animation, retro palettes, dithering, shading, and spritesheet export.
Low-VRAM LoRA Training for Krea-2.
Free, local-first voice assistant for your desktop. Dictation, writing, and an AI assistant, all by voice. Runs on-device, cross-platform, open source.
Agents orchestrating other agents.
Make Claude Code a .NET 10 Expert.
AI Model Demand Offloading Allocator.
Bridge Feishu/Lark to AI coding CLIs — Claude Code, Codex, Gemini, OpenCode… every DM, group or topic spawns its own live-streaming CLI session
Sandbox any AI agent in seconds - zero setup, zero latency.
Run Kimi K3, a 2.8T-parameter Mixture-of-Experts LLM, on a single Apple Silicon Mac. Streams MXFP4 experts on demand over HTTP into a local disk cache — fused NEON kernels, Metal/MPS compute, exact reproducible decoding, and an OpenAI-compatible API server for local chat and coding agents.
ExploitGym is a large-scale, realistic benchmark built from real-world vulnerabilities designed to evaluate AI agents' ability to develop exploits.
[ICML 2026 Spotlight] Official repository of Chamaileon: Cross-Context Binder Design with Contextualized Modeling and Mixed Sampling
Turn one topic into a finished Vox-style paper-collage explainer/ad video — automated end to end on Atlas Cloud + ffmpeg. An agent skill.
Visualize GitHub repositories, explore project structure, and find the right files for any issue using AI