Interactive research previews

Explore DiffSpace Demos

A curated gallery of interactive demos from our work on diffusion language models, speculative decoding, in-place editing, and vision-language-action systems.

All demos

05 / LIVE
01 LLM

DiffSpace Chat

A chat interface backed by a diffusion language model — history, settings, and streaming-style generation in a clean conversational UI.

Chat Diffusion LLM
Open
02 Speculative

DFlash

Block-diffusion speculative decoding — a visual walkthrough of how diffusion-style drafts accelerate autoregressive inference, with live accept/reject animation.

Diffusion Inference
Open
03 Tool Use

Parallel Tool Calling

Diffusion LLM × parallel tool calls — shows how a non-autoregressive decoder emits multiple tool invocations in one pass, instead of sequentially.

Diffusion LLM Agents
Open
04 Editing

In-place Editing

Document-level diffusion edits with global consistency — local edits, conclusion→intro propagation, and multi-span infilling, animated step by step.

Infilling Consistency
Open
05 VLA

AR vs Diffusion Policy

Side-by-side VLA rollouts — autoregressive vs diffusion action policies on the same tasks, rendered from real robot trajectories.

Robotics Policy
Open