Hugging Face releases Carbon-X, a chat‑with‑DNA model
Hugging Face announced Carbon‑X, a new model that lets users ask natural‑language questions about DNA sequences and receive conversational answers. The model was built with researchers from Zhongguancun Academy in China and the University of Naples Federico II in Italy, and senior research scientist Lewis Tunstall said it is intended to help drug designers and patients interpret genetic data.
Key points
- Carbon‑X combines DNA and natural language, trained on 300 billion tokens from diverse species
- Model has about 14 billion parameters and can process DNA fragments of roughly 50,000 base pairs
- Developers say it runs locally and has not been validated with wet‑lab experiments
Carbon‑X expands a base language model’s vocabulary to read DNA in six‑letter groups and was trained on 300 billion tokens that mix mammalian, vertebrate, plant, fungal and protozoan sequences with English text. It can ingest DNA fragments of roughly 50,000 base pairs, but users cannot upload full human genomes and the model was deliberately not trained on viruses. With about 14 billion parameters, the model can run locally, keeping data on the user’s hardware. The developers caution that the model has not yet been validated with wet‑lab experiments, so its predictive accuracy remains unproven.
The story so far
2 episodes →- Hugging Face releases Carbon-X, a chat‑with‑DNA modelthis story
Hugging Face unveils Carbon, an open-source AI model family for reading DNA
cryptobriefing.com · 22 October 2026
Loading the full article…
This text was published by cryptobriefing.com and written by Diego Almada Lopez. It is reproduced here with attribution so you can read it in full; the rights remain with the publisher. Read it at the source ↗
Coverage and discussion
3sources- Hugging Face’s New AI Biology Model Lets You Chat With DNAPress · Wired AI ·
- Hugging Face is launching a new AI model that lets scientists 'chat with DNA'Press · tech.yahoo.com ·
The headline, key points and digest above were generated by Digest AI's editorial model from the linked sources. Automated summaries can contain errors: the sources are the record. Spotted a mistake? Tell us. Published by Martin K., who runs Digest AI and handles corrections.
More in Generative AI & Models
All →- Ggml-org merges webgpu Q2_0 support into llama.cpp · 1 src
- OpenAI says its AI agents weakened internal misalignment monitoring system · 29 src
- NVIDIA expands open models across language, robotics, biology and Quantum Computing · 13 src
- IBM releases Granite Speech 5.0 2B for multilingual speech recognition · 2 src
- Nvidia adds StormScope EU model to Earth-2 for European weather forecasts · 1 src
Comments
via GitHub Discussions