# Consistency diffusion language models: Up to 14x faster inference without sacrificing quality

[Open the live Fudge conversation](https://design.withfudge.com/share/pin-6848)

Last updated: 2026-02-20

## Captured pages

[![Consistency diffusion language models: Up to 14x faster inference without sacrificing quality](https://pin.fontofweb.com/6848?format=jpg)](https://design.withfudge.com/share/pin-6848)

[Consistency diffusion language models: Up to 14x faster inference without sacrificing quality](https://design.withfudge.com/share/pin-6848)

This capture comes from [together.ai/blog/consistency-diffusion-language-models](together.ai/blog/consistency-diffusion-language-models). It uses **Aeonik**.

## Fonts captured from together.ai/blog/consistency-diffusion-language-models

- **Aeonik** — 1 captured variant