Geneticists ranking variants, including in non-coding DNA, get a no-code portal, though pricing and usage limits haven't been published.
Kenning's take
What it does
AlphaGenome Atlas is a 1-petabyte dataset made by running Google DeepMind's AlphaGenome model over all 9 billion possible single-letter changes in the human genome. It gives each change an AlphaGenome Variant Impact (AVI) score that combines AlphaGenome and AlphaMissense predictions, so researchers can rank which mutations are most likely to matter, including in the 98% of DNA that does not code for proteins. It is available through a web portal that needs no coding, the AlphaGenome API, and as a skill in Google Antigravity.
- Predictions for all 9 billion single-nucleotide variants
- AlphaGenome Variant Impact (AVI) score
- Covers coding and non-coding DNA
- No-code web portal
- AlphaGenome API and Google Antigravity skill
- Catalogue of over 2,500 recurrent DNA sequences
Who it suits. Geneticists, clinical researchers and biologists studying rare diseases and the genetics of common traits.
Key facts
- Category
- AI models
- Made by
- Google DeepMind
- Released
- 8 September 2026
- How to get it
- web portal at alphagenome.google/atlas (Google sign-in), the AlphaGenome API, and a skill in Google Antigravity
- Works on
- Web, API
- In the UK
- Yes
FAQ
When was AlphaGenome Atlas released?
Google DeepMind announced it on 8 September 2026.
How do I use AlphaGenome Atlas?
Through a web portal that needs no coding, through the AlphaGenome API, or as a skill in Google Antigravity.
What is the AVI score?
The AlphaGenome Variant Impact score is a single number combining AlphaGenome and AlphaMissense predictions to rank how much a DNA change is likely to matter.
Facts from the maker's own pages and the sources listed. Checked 2026-10-11.
Similar tools
ACE-Step 1.5
Open-source music generation model that writes full songs with lyrics and runs on ordinary consumer GPUs.
Amazon Nova 2
Amazon's second family of Nova AI models, covering reasoning, speech conversation and combined text-and-image generation.
Cisco Antares
Cisco's small open-weight models that find where known security vulnerabilities sit in a codebase, running locally.
QuiverAI Arrow 2
QuiverAI's model for generating, vectorising and animating SVG graphics, released with a higher-end Telos version.