OpenModelsHQ

Stability AI

London, UK

Stability AI is based in London, UK. We track 15 open-weight models from it, and the newest is stable-audio-3-optimized, released 18 May 2026.

Models tracked
15
LLM 1 · Image 7 · Video 3 · Speech 4
Newest release
stable-audio-3-optimized
18 May 2026
Country
UK
Releases timeline
15 Stability AI releases from 25 Jul 2023 to 18 May 2026. The largest is stable-diffusion-3.5-large at 8.1B parameters. Dot area follows total parameters; a ring means the size is not published.
Releases timeline: data
ModelCategoryReleasedParameters
stable-audio-3-optimizedSpeech18 May 2026Not published
stable-audio-3-small-sfxSpeech17 May 2026568M
stable-audio-3-mediumSpeech17 May 20262.3B
stable-virtual-cameraVideo19 Feb 20251.3B
stable-diffusion-3.5-mediumImage29 Oct 20242.5B
stable-diffusion-3.5-largeImage22 Oct 20248.1B
stable-diffusion-3-medium-diffusersImage12 Jun 20242.1B
stable-audio-open-1.0Speech24 May 20241.2B
sd-turboImage27 Nov 2023866M
sdxl-turboImage27 Nov 20232.6B
stablelm-zephyr-3bLLM21 Nov 20232.8B
stable-video-diffusion-img2vid-xtVideo20 Nov 20231.5B
stable-video-diffusion-img2vidVideo20 Nov 20231.5B
stable-diffusion-xl-refiner-1.0Image26 Jul 20232.3B
stable-diffusion-xl-base-1.0Image25 Jul 20232.6B

History coming soon.

Release lineage

Newest first
  1. stable-audio-3-optimized
  2. stable-audio-3-small-sfx568M
  3. stable-audio-3-medium2.3B
  4. stable-virtual-camera1.3B
  5. stable-diffusion-3.5-medium2.5B
  6. stable-diffusion-3.5-large8.1B
  7. stable-diffusion-3-medium-diffusers2.1B
  8. stable-audio-open-1.01.2B
  9. sd-turbo866M
  10. sdxl-turbo2.6B
  11. stablelm-zephyr-3b2.8B
  12. stable-video-diffusion-img2vid-xt1.5B

Models from Stability AI

Compare in the table

Largest first. The last line on each tile is the memory of the smallest device that fits the model at 4-bit.