Intel / post

LLaDA2.1-flash goes live on ZenMux for hosted API access

Source: x.com
Date: Published Mar 16, 2026

⚡️ 892 tokens/s — our 100B diffusion LLM, LLaDA2.1-flash, is now live on @ZenMuxAI! With Token Editing, LLaDA 2.1 goes from research breakthrough to production-ready speed. Diffusion models just got real. Try it via API or Chat 👇 https://t.co/8ObarWTPio #LLaDA #ZenMux #AI #dLLM https://t.co/zUUROipUAs

Mar 16, 2026 View on X

Context

A number for a new model architecture means little until you can actually call it the way you'd call any other API. Ant Open Source says its 100B-parameter diffusion model, LLaDA2.1-flash, is now live on the ZenMux platform through both a chat interface and an API, at a stated 892 per second. That makes the token-editing speed claim behind LLaDA2.1, first reported at the model's launch, something a developer can test against their current stack without provisioning their own GPUs.

Terms in this piece · Glossary

benchmark — A standard public test set for comparing AI models — the shared scoreboards behind every "model X beats model Y" claim.
token — The chunk of text a model reads and writes in — roughly three-quarters of a word — and the unit AI usage is billed in.

Read the source x.com