Hugging Face logo
Hugging Face for Startups

Hugging Face for Startups

Offers Hugging Face startup credits for teams building model-based products and experimentation.

OpenHugging FaceGlobalDeep-tech · core fit

Eligibility · Global — open worldwide

Hugging Face for Startups refers collectively to the in-kind support channels operated by Hugging Face, the New York City-based open-source AI platform company founded in 2016 with a mission to democratize machine learning. The primary first-party program is the Community GPU Grants track, which allocates in-kind GPU compute — defaulting to ZeroGPU (Nvidia RTX Pro 6000 Blackwell, up to 96 GB GPU memory) or A10G in some cases — to qualifying public Spaces projects on the Hugging Face Hub. Applications are submitted directly through a Space's hardware settings. A secondary channel routes Pro-account credits (six months of Hugging Face Pro free) through partner accelerators and startup-perks marketplaces such as FounderPass and XRaise; these are not a structured first-party grant program with a public application page.

For the Community GPU Grants track, awards are in-kind compute with an implied monthly retail equivalent of roughly $720–$1,800 for ZeroGPU or A10G usage at continuous utilization. No cash is disbursed. Grants are temporary and may be removed if usage is low; duration is case-by-case with no fixed term. Eligibility favors academic research projects, open-source community demos, and projects with demonstrated traction (visits, GitHub stars, published papers). There are no geographic restrictions and no funding-stage requirements, making the track accessible to individuals, academic teams, and indie developers alongside startups. Structured fields indicate the award range at $500 to $100,000, reflecting the wide variance across different support tiers.

Teams seeking meaningful GPU allocation for applied AI or ML research should apply via the Space hardware settings on huggingface.co and demonstrate a working, publicly accessible project with clear community value. Startups seeking Pro credits via accelerator partnerships should look for Hugging Face listed in their accelerator's perks marketplace rather than applying directly through a Hugging Face grant portal.

Provides in-kind GPU compute credits for public AI and ML Spaces projects on the Hugging Face Hub, with additional Pro-account credits available to startups via partner accelerator programs.

CycleiHow often this grant runs — e.g. annually, on a rolling basis, or a one-off call.Rolling
Next deadlineiThe next date applications are due. Rolling means you can apply any time.Rolling
Decision timeiTypical time from the deadline to the funder's decision.
Project durationiHow long the funded work is expected to run.
Award typeiThe form of funding — grant, equity, loan, tax credit, etc.Credits
Match fundingiThe share of project costs you must cover yourself. 0% = fully funded.0%
Funding pooliThe total budget available across all awards in this round.

Sign up free to see the funding breakdown

Sign up free to see the industries in scope

Sign up free to see the full eligibility

Sign up free to see how to apply

Last verified: 29 Jun 2026Source: huggingface.co