Carbon-X
Carbon-X is a genomics & dna models tool, first public in 2026. Hugging Face's 14.8B DNA-and-English chat model built on Qwen3-14B: paste up to ~50 kbp of DNA and ask about variant effects, regulatory regions, organism group or sequence continuation, with explanations.
| Category | Genomics & DNA models |
|---|---|
| Task | Variant effect (benign vs. pathogenic), regulatory region identification, organism-group classification, sequence continuation with explanations |
| Interface | Chat model via Transformers, vLLM, llama.cpp (GGUF) |
| Website | huggingface.co/HuggingFaceBio/Carbon-X |
| Released | 2026-10-22 (source) |
| Code license | Apache-2.0 |
| Weights license | Apache-2.0 (weights and GGUF quantizations) |
| What this allows | Permissive (commercial OK) Use, modify and ship it inside closed-source commercial products; keep the license text and copyright notices. What do these mean? |
| License source | huggingface.co/HuggingFaceBio/Carbon-X#license · checked 2026-10-22 |
| Pricing | Open source |
| Tags | #dna-model#variant-effect#llm#chat |
| Last checked | · links verified |
Can I use Carbon-X commercially?
Yes. The recorded license is permissive, so closed-source commercial use is allowed if you keep the notices.
- Code: Apache-2.0
- Weights: Apache-2.0 (weights and GGUF quantizations)
- Source: huggingface.co/HuggingFaceBio/Carbon-X#license · license checked 2026-10-22
What the license types mean · not legal advice.
Specs
| Input length / context | ~50 kbp of DNA (trained on 8,192-token windows, ≈49 kbp); 40,960-token context |
|---|---|
| Resolution | 6-mer DNA tokens (hybrid tokenizer: Qwen3 BPE for text) |
| Species | Eukaryotes (vertebrates incl. mammals, invertebrates, plants, fungi, protozoa) and prokaryotes; eukaryote-infecting viruses excluded |
| Tasks | Variant effect (benign vs. pathogenic), regulatory region identification, organism-group classification, sequence continuation with explanations |
| Interface | Chat model via Transformers, vLLM, llama.cpp (GGUF) |
Compare with…
Pick 1–3 other Genomics & DNA models tools to compare side by side with Carbon-X.
Related
Predicts gene expression from sequence with long-range interactions.
Genomics & DNA modelsOpen sourcePermissive (commercial OK)#expression
Genome language model across all domains of life with 1M-token context.
Genomics & DNA modelsOpen sourcePermissive (commercial OK)#dna-lm
DeepMind model predicting regulatory effects of DNA variants at base resolution.
Genomics & DNA modelsNo cost (research use)Mixed (code vs weights differ)#regulatory