Stanford’s Le Cong lab and collaborators used Nebius AI Cloud GPU clusters to develop CRISPR-GPT, an AI assistant for gene-editing experiments. Its architecture combines an LLM planner, task executors, user-proxy agents and external bioinformatics tools. Nebius describes training CRISPR-Llama3 on 11 years of scientific-forum discussions and experimenting on an eight-H100 node. The paper describes curating approximately 4,000 discussion threads into over 3,000 question-and-answer pairs to fine-tune an eight-billion-parameter Llama3-instruct model. Parallel model testing reportedly saved several months of development time. Preconfigured libraries and drivers allowed workload deployment without infrastructure setup.
Nebius’s podcast representative also describes training and running the project’s models on Nebius infrastructure. The team developed Gene-editing bench, a 288-case evaluation framework covering experiment planning, guide RNA design, delivery-method selection and Q&A. Human experts supplied ground-truth annotations for benchmarking against existing tools and general-purpose LLMs. Nebius reports that a novice PhD student achieved approximately 80% editing efficiency across four targets on the first attempt. An undergraduate achieved 56.5% and 90.2% efficiency on two separate targets. Nebius lists the paper among research supported by its grants programme.
Le Cong later described collaboration with Nebius on LabOS’s AI and XR laboratory research and Nebius support for the LabSuperVision benchmark.
Nebius says the CRISPR-GPT team used Nebius AI Cloud GPU clusters for architecture experiments and CRISPR-Llama3 fine-tuning on 11 years of scientific-forum discussions. Parallel model testing saved several months of development time. Preconfigured libraries and drivers enabled deployment without infrastructure setup. The team’s Gene-editing bench includes 288 cases covering experiment planning, guide RNA design, delivery-method selection and Q&A, with human-expert ground-truth annotations for benchmarking against existing tools and general-purpose LLMs.
Nebius Stanford CRISPR-GPT research story5 Aug 2025 · Nebius · Web pageNebius says the CRISPR-GPT team used Nebius AI Cloud GPU clusters for architecture experiments and CRISPR-Llama3 fine-tuning on 11 years of scientific-forum discussions. Parallel model testing saved several months of development time. Preconfigured libraries and drivers enabled deployment without infrastructure setup. The team’s Gene-editing bench includes 288 cases covering experiment planning, guide RNA design, delivery-method selection and Q&A, with human-expert ground-truth annotations for benchmarking against existing tools and general-purpose LLMs.
CRISPR-GPT for agentic automation of gene-editing experiments | Nature Biomedical Engineering Close banner Close banner
14 Oct 2026 · www.nature.com · Web page
The CRISPR-GPT authors acknowledge supercomputing resources provided by Nebius Group N.V. and the Stanford Genetics Bioinformatics Service Center. They describe curating approximately 4,000 scientific discussion threads into over 3,000 question-and-answer pairs to fine-tune an eight-billion-parameter Llama3-instruct model.
CRISPR-GPT for agentic automation of gene-editing experiments | Nature Biomedical Engineering Close banner Close banner14 Oct 2026 · www.nature.com · Web pageThe CRISPR-GPT authors acknowledge supercomputing resources provided by Nebius Group N.V. and the Stanford Genetics Bioinformatics Service Center. They describe curating approximately 4,000 scientific discussion threads into over 3,000 question-and-answer pairs to fine-tune an eight-billion-parameter Llama3-instruct model.
Cloud for Scientific AI and Healthcare
14 Oct 2026 · nebius.com · Web page
Nebius says Stanford-led CRISPR-GPT used its infrastructure for rapid model screening and fine-tuning. It reports 80–90% first-attempt editing efficiency for novice researchers and undergraduate onboarding in one day.
Cloud for Scientific AI and Healthcare14 Oct 2026 · nebius.com · Web pageNebius says Stanford-led CRISPR-GPT used its infrastructure for rapid model screening and fine-tuning. It reports 80–90% first-attempt editing efficiency for novice researchers and undergraduate onboarding in one day.
Why AI Medical Research Needs Vast Computing Power and Where Nebius AI Cloud Fits In - Beyond Longevity
14 Oct 2026 · beyond-longevity.co.uk · Web page
The Nebius representative says Nebius supplied H100 compute for Le Cong’s Stanford team to develop CRISPR-GPT after its internal computing resources became limiting.
Why AI Medical Research Needs Vast Computing Power and Where Nebius AI Cloud Fits In - Beyond Longevity14 Oct 2026 · beyond-longevity.co.uk · Web pageThe Nebius representative says Nebius supplied H100 compute for Le Cong’s Stanford team to develop CRISPR-GPT after its internal computing resources became limiting.
LabOS at Jensen Huang's Keynote at GTC DC
30 Oct 2025 · www.linkedin.com · Web page
Le Cong describes collaboration with Nebius and other partners on LabOS’s AI and XR laboratory research. He also announces the LabSuperVision benchmark with support from Nebius, NVIDIA and Viture.
LabOS at Jensen Huang's Keynote at GTC DC30 Oct 2025 · www.linkedin.com · Web pageLe Cong describes collaboration with Nebius and other partners on LabOS’s AI and XR laboratory research. He also announces the LabSuperVision benchmark with support from Nebius, NVIDIA and Viture.
CRISPR-GPT is an academic research project developed by researchers at Stanford, Princeton and collaborating institutions. Its AI assistant helps biomedical researchers plan and analyze CRISPR gene-editing experiments, including selecting editing systems, designing guide RNAs, choosing delivery methods, drafting protocols and interpreting results. It supports knockout, base editing, prime editing and epigenetic editing, with guided workflows and question-answering interfaces for researchers of different experience levels. The team provides a limited developer preview through its official GitHub repository.
Stanford’s Le Cong lab and collaborators used Nebius AI Cloud GPU clusters to develop CRISPR-GPT, an AI assistant for gene-editing experiments. Its architecture combines an LLM planner, task executors, user-proxy agents and external bioinformatics tools. Nebius describes training CRISPR-Llama3 on 11 years of scientific-forum discussions and experimenting on an eight-H100 node. The paper describes curating approximately 4,000 discussion threads into over 3,000 question-and-answer pairs to fine-tune an eight-billion-parameter Llama3-instruct model. Parallel model testing reportedly saved several months of development time. Preconfigured libraries and drivers allowed workload deployment without infrastructure setup.
Nebius’s podcast representative also describes training and running the project’s models on Nebius infrastructure. The team developed Gene-editing bench, a 288-case evaluation framework covering experiment planning, guide RNA design, delivery-method selection and Q&A. Human experts supplied ground-truth annotations for benchmarking against existing tools and general-purpose LLMs. Nebius reports that a novice PhD student achieved approximately 80% editing efficiency across four targets on the first attempt. An undergraduate achieved 56.5% and 90.2% efficiency on two separate targets. Nebius lists the paper among research supported by its grants programme.
Le Cong later described collaboration with Nebius on LabOS’s AI and XR laboratory research and Nebius support for the LabSuperVision benchmark.
Nebius says the CRISPR-GPT team used Nebius AI Cloud GPU clusters for architecture experiments and CRISPR-Llama3 fine-tuning on 11 years of scientific-forum discussions. Parallel model testing saved several months of development time. Preconfigured libraries and drivers enabled deployment without infrastructure setup. The team’s Gene-editing bench includes 288 cases covering experiment planning, guide RNA design, delivery-method selection and Q&A, with human-expert ground-truth annotations for benchmarking against existing tools and general-purpose LLMs.
Nebius Stanford CRISPR-GPT research story5 Aug 2025 · Nebius · Web pageNebius says the CRISPR-GPT team used Nebius AI Cloud GPU clusters for architecture experiments and CRISPR-Llama3 fine-tuning on 11 years of scientific-forum discussions. Parallel model testing saved several months of development time. Preconfigured libraries and drivers enabled deployment without infrastructure setup. The team’s Gene-editing bench includes 288 cases covering experiment planning, guide RNA design, delivery-method selection and Q&A, with human-expert ground-truth annotations for benchmarking against existing tools and general-purpose LLMs.
CRISPR-GPT for agentic automation of gene-editing experiments | Nature Biomedical Engineering Close banner Close banner
14 Oct 2026 · www.nature.com · Web page
The CRISPR-GPT authors acknowledge supercomputing resources provided by Nebius Group N.V. and the Stanford Genetics Bioinformatics Service Center. They describe curating approximately 4,000 scientific discussion threads into over 3,000 question-and-answer pairs to fine-tune an eight-billion-parameter Llama3-instruct model.
CRISPR-GPT for agentic automation of gene-editing experiments | Nature Biomedical Engineering Close banner Close banner14 Oct 2026 · www.nature.com · Web pageThe CRISPR-GPT authors acknowledge supercomputing resources provided by Nebius Group N.V. and the Stanford Genetics Bioinformatics Service Center. They describe curating approximately 4,000 scientific discussion threads into over 3,000 question-and-answer pairs to fine-tune an eight-billion-parameter Llama3-instruct model.
Cloud for Scientific AI and Healthcare
14 Oct 2026 · nebius.com · Web page
Nebius says Stanford-led CRISPR-GPT used its infrastructure for rapid model screening and fine-tuning. It reports 80–90% first-attempt editing efficiency for novice researchers and undergraduate onboarding in one day.
Cloud for Scientific AI and Healthcare14 Oct 2026 · nebius.com · Web pageNebius says Stanford-led CRISPR-GPT used its infrastructure for rapid model screening and fine-tuning. It reports 80–90% first-attempt editing efficiency for novice researchers and undergraduate onboarding in one day.
Why AI Medical Research Needs Vast Computing Power and Where Nebius AI Cloud Fits In - Beyond Longevity
14 Oct 2026 · beyond-longevity.co.uk · Web page
The Nebius representative says Nebius supplied H100 compute for Le Cong’s Stanford team to develop CRISPR-GPT after its internal computing resources became limiting.
Why AI Medical Research Needs Vast Computing Power and Where Nebius AI Cloud Fits In - Beyond Longevity14 Oct 2026 · beyond-longevity.co.uk · Web pageThe Nebius representative says Nebius supplied H100 compute for Le Cong’s Stanford team to develop CRISPR-GPT after its internal computing resources became limiting.
LabOS at Jensen Huang's Keynote at GTC DC
30 Oct 2025 · www.linkedin.com · Web page
Le Cong describes collaboration with Nebius and other partners on LabOS’s AI and XR laboratory research. He also announces the LabSuperVision benchmark with support from Nebius, NVIDIA and Viture.
LabOS at Jensen Huang's Keynote at GTC DC30 Oct 2025 · www.linkedin.com · Web pageLe Cong describes collaboration with Nebius and other partners on LabOS’s AI and XR laboratory research. He also announces the LabSuperVision benchmark with support from Nebius, NVIDIA and Viture.
CRISPR-GPT is an academic research project developed by researchers at Stanford, Princeton and collaborating institutions. Its AI assistant helps biomedical researchers plan and analyze CRISPR gene-editing experiments, including selecting editing systems, designing guide RNAs, choosing delivery methods, drafting protocols and interpreting results. It supports knockout, base editing, prime editing and epigenetic editing, with guided workflows and question-answering interfaces for researchers of different experience levels. The team provides a limited developer preview through its official GitHub repository.
