Been building the demos behind model launches you've probably seen — quietly, with some great model teams. So many possibilities, poked and pushed and tested. Sharing the work here now. 🧵 — Mike Luan, e01

Claude Fable 5 one shotted this chess playing robot sim sim->real under 20 minutes. GPT5.5 xhigh did too, but then Fable went physical...

A Lab note for Step 3.7 Flash launch. -- When Flash models bring speed, cost, and intelligence into the “usable” range all at once, the way intelligence is supplied changes structurally. -- blog.e01.ai/beyond-speed-the…

Adding to the "Sampling the Model" Track, Parallel requests to Step-3.7-Flash with cluster of persona, voting in realtime.

With launch of Step 3.7 Flash by @StepFun_ai , we explored multiple new use-cases with "Flash Class" models. The balance of speed, cost and quality gives life to new way to interact with the model. This demo shows possibility to "Sample" the model in realtime, revealing layer of intelligence instead of a single "chat". Write up coming soon.

lean & fast models fast infra do make a difference. in our latest experiment {consensus}, 50~100 persona (agents) are generated to answer one question. turning "query" into "sampling", to potentially "extract" more from the model offerings. e01.ai/consensus/ repo: github.com/e01-ai/consensus

We've added support for more model providers (openrouter | fireworks | gemini) to vcanvas and pushed source to github. Grab the source here: github.com/e01-ai/vcanvas Play with more providers here: e01.ai/vcanvas/ Enjoy!

We'll add more models / providers to vcanvas playground soon. e01.ai/vcanvas

Introducing GLM-5V-Turbo: Vision Coding Model - Native Multimodal Coding: Natively understands multimodal inputs including images, videos, design drafts, and document layouts. - Balanced Visual and Programming Capabilities: Achieves leading performance across core benchmarks for multimodal coding, tool use, and GUI Agents. - Deep Adaptation for Claude Code and Claw Scenarios: Works in deep synergy with Agents like Claude Code and OpenClaw. Try it now: chat.z.ai API: docs.z.ai/guides/vlm/glm-5v-… Coding Plan trial applications: docs.google.com/forms/d/e/1F…

Collaborating with @Zai_org on early exploration around GLM-5V-Turbo (just launched), a strong, fast & multi-modal model. Richer modality = more creative freedom. We explored the interface side to enable it: a canvas... makes visual interaction fluid — draw, collage, iterate, all in one space. Pair that with 5V-Turbo's coding capability, you get sketch-to-running-code in seconds. e01.ai/vcanvas/

Prompts were used to generate programs. Now prompts are becoming programs themselves. We traced the evolution — from magic words that decompress into entire knowledge structures, to prompts that loop, to 39 lines that self-assemble an agent team from scratch. From equivalent length to orders of complexity. Our observation. A short read: blog.e01.ai/prompts-are-more…

GLM-5, Gameboy and Long-Task Era → 700 tool calls, 800 context handoffs, and a single agent running for over 24 hours. blog.e01.ai/glm5-gameboy-and…