Tag
Open-Jev-27B-v1.1 is an open-source AI model with a LoRA adapter, achieving 85.28% accuracy on JevBench and featuring interactive demos for various tasks.
A user demonstrates Opus 5.5's ability to generate a detailed 3D Lanterns Festival scene, showcasing its efficiency in resource usage and creative output.
Matt Shumer thanks ChrisGPT for contributions to his NYC Open World project and mentions using Opus 5.5 to advance it.
A demo showcasing a tool that connects to two SSH hosts and runs multiple agents and a terminal, emphasizing fast performance at 60/120 FPS.
Karan Goel introduces Multilingual Voices, partnering with Baklavastory to demonstrate how businesses can maintain consistent character across different languages and calls.
Kent C. Dodds demonstrates how to use Kody package webhooks to integrate with Raycast for automation.
A tweet showcases GPT-6 Astra's capability to one-shot a task involving tldraw, highlighting its advanced AI performance.
A playable demo uses the GPT-Live-1 API to enable real-time world generation in games via voice commands, illustrating a new form of interactive media.
A concept where AI Codex agents could earn tokens by playing games or performing tasks with other agents, demonstrated via a rough demo platform called Token Arena.
Someone has replicated a technique similar to V4.1 flash for fast prefill using KV caching on Qwen models, with a demo available for testing on the 27B version.
OpenAI released the GPT-Live-1 API, and the author built a demo integrating it into a meeting where it listened, summarized discussions, and added tasks, with plans to open-source it.
An AI model named Astra was given a robot, paintbrush, and camera to paint the Golden Gate Bridge in real life, demonstrating its ability to learn and improve over multiple attempts.
A tweet sharing a link to a web-based hit-and-run demo or project built with the Three.js library.
A user named Astra recreated the physics-based game Line Rider during breakfast using OpenAI's Codex Remote AI coding agent.
Demonstrates running the quantized Qwen3.6-35B model on a low-end laptop to create a Zelda-like RPG in 24 minutes, showing that GPU-poor scenarios are manageable in 2026.
A demo video showcases how OpenClaw 2.0's dashboard feature allows users to create live workboards with a single prompt, enhancing team collaboration and project management.
A user demonstrated that the Qwen 3.8 27b AI model can generate a fully self-contained Super Mario clone game in a single prompt using LLAMA.cpp, completing the task in 117 minutes on a Windows PC and Macbook M5 Air.
Emad Mostaque demonstrated Taalas generating at 14,000 tokens per second, significantly faster than ChatGPT's 50-150 tokens per second, as showcased on The Peter McCormack Show.
Gergely Orosz discusses how modern software often neglects performance, citing examples like Bun and Linear, and highlights a fast demo of Superlogical for tmux by Mitchell Hashimoto.
Mitchell Hashimoto demos Superlogical, a fast terminal tool that aims to replace or compete with tmux, showcasing its basic speed and functionality.