Schmidt Sciences' Unconventional Compute RFP funds research on hardware fundamentally different from conventional CPUs and GPUs, co-designed with training and inference methods suited for their constraints. Open to universities and nonprofit research institutions globally, this pilot program seeks evidence that unconventional hardware can solve real-world problems beyond benchmark metrics.

Track I awards $50,000-$150,000 for 6-12 month projects; Track II awards $150,000-$750,000 for 12-18 months. The lightweight application requires only a short online form and a five-page project narrative. Deadline is April 30, 2026.

2026 Unconventional Compute RFP - Schmidt Sciences 2026 Unconventional Compute RFP Opens Feb 13 2026 12:00 AM (EST) Deadline Apr 30 2026 11:59 PM (EDT) Request for Proposals: Unconventional Compute This is a pilot program. We are seeking evidence that unconventional compute hardware can tackle substantial, real-world problems that go beyond toy metrics.

If this pilot uncovers compelling evidence, it may unlock a significantly larger investment in this space. This is a lightweight application. There are no letters of support, no spreadsheets to be filled.

The application consists of a short online form and a (less than) 5-page project narrative pdf. We're interested in what you've built and where you want to take it—not your ability to write long proposals.

Schmidt Sciences seeks to fund research that can catalyze hardware fundamentally different from today’s CPU–GPU paradigm, and the training/inference methods co-designed to operate under the constraints imposed by such hardware: low numerical precision, device noise, drift, non-idealities etc. We are looking for proposals grounded in the creation of hardware that will solve specific, well-defined technological problems where conventional compute underperforms.

This request is open to universities and non-profits globally. You are building or designing compute hardware that is NOT based on conventional CPUs or GPUs. You have a real-world task in mind where your hardware could eventually outperform GPUs.

You are at a university or nonprofit research institution anywhere in the world. This RFP is not for you if: Your work is primarily about improving GPU-based architectures You don't yet have a target application in mind You are based at a for-profit institution.

April 30, 2026, 11:59PM Anywhere on Earth How to Apply Apply via SurveyMonkey Funding Tracks Track I: $50,000 - $150,000 ( 6 - 12 months) Track II: $150,000 - $750,000 (12 - 18 months) Estimated Number of Awards March 12, 2026, 10-11am EST. Register here March 19, 2026, 3-4pm EST. Register here April 7, 2026, 3-4pm EST.

Register here This request is open to universities and non-profits globally. aiforscience@schmidtsciences. org Frequently Asked Questions Indirect Costs Policy Indirect costs must not exceed 10% for the entire project.

Is there compute hardware that can outperform GPUs (latency, energy, throughput, etc.) in a technologically important niche? AI performance has been driven by scaling models and compute on CPU–GPU stacks, but the energy, latency, and cost of moving data is becoming the bottleneck. The next wave of computing need not look like a Transformer running on a GPU.

We believe that by exploiting novel physics and chemistry— for instance through optics, thermodynamics, or polymers—we can build hardware that solves specific, high-value problems that are intractable or inefficient with conventional compute.

This RFP seeks to identify and fund research that can catalyze hardware fundamentally different from today’s CPU–GPU paradigm, and the training/inference methods co-designed to operate under the constraints imposed by such hardware: low numerical precision, device noise, drift, and non-idealities.

We are looking for proposals grounded in the creation of hardware that will solve specific, well-defined technological problems where conventional compute underperforms. For example, some well-defined technological problems could include imaging flow cytometry, RF signal processing and high throughput industrial vision. We welcome applications across all domains.

We particularly encourage applications from teams with existing hardware, preliminary data, or demonstrated capabilities where additional resources would help drive towards more ambitious work. Successful projects that demonstrate valid physics and a credible demo will be considered for follow-on funding.

Funded teams will be invited to a convening of researchers and industry leaders in this space (anticipated Summer 2026), creating opportunities for collaboration and visibility. Applicants can apply for one of two funding track: Track I: $50,000 - $150,000 (6-12 months) Track I funds simulations, preliminary experiments, or proof-of-concept work that establishes the feasibility of a substantial prototype.

Successful Track I projects will produce evidence that a specific unconventional compute approach can address a real-world use case, with quantified performance projections.

Track I is appropriate if you: Have a credible approach but need to demonstrate feasibility before building hardware Need simulation or small-scale experimental data to validate performance claims Can produce preliminary evidence within 6-12 months Track II: $150,000 - $750,000 (12-18 months) Track II funds the creation of substantial prototypes that perform real-world functions.

Track II applicants must demonstrate that feasibility has already been established—through prior work, published results, or existing hardware. Track II is appropriate if you: Have existing hardware or strong preliminary data demonstrating feasibility. Are ready to build or scale a prototype for a specific application.

Can show meaningful progress within 12-18 months. We will fund projects focused on performance on a specific, substantial task(s).

In-scope approaches (illustrative, not exhaustive): Optical/photonic processing RF/microwave photonics and native RF-domain processing Resistive and mixed-signal compute Neuromorphic, stochastic/approximate computing In-scope task types (illustrative, not exhaustive): Real-time video / optical preprocessing and filtering at the sensor High-throughput microscopy classification/triage Always-on embedded sensing and event detection (industrial monitoring, wearables) Any other clearly defined application where a GPU is disadvantaged.

What we will not fund (generally) “Interesting hardware” without a compelling use case Toy use cases (e.g. MNIST digit recognition task) Proposals where the primary bottleneck is chemistry/unknown materials behavior Biology-based compute (e.g. DNA computing) Submissions will be assessed by internal and external reviewers with expertise relevant to your proposal topic.

During the assessment process, we may reach out to applicants to clarify portions of their proposal. Assessment criteria include: Problem Significance : Does the proposal identify a well-defined technological bottleneck (e.g., "ultra high speed industrial imaging") rather than a generic goal (e.g., "make AI faster")? Execution capability : team, facilities and proven ability.

What has the team already demonstrated? Is there existing momentum? Well-characterized physical phenomena and materials : We deprioritize proposals whose success depends heavily on the discovery of novel materials or chemistries.

Scalability : What is the likelihood that the proposed research, in the long term, will create new technology? Documents should be submitted through the SurveyMonkey Apply portal. Below is a list of the items you will need to submit: Name, email and title of the person submitting the proposal.

Institution/Recipient of Fund. Choice of Funding Track (I or II) Total Project Budget (USD) Project Start Date (MM/YY) Please upload the CV of the lead PI (mandatory) and other team members (optional) (Must be uploaded as a PDF.) Please provide a short abstract of your project written for a non-specialist audience (< 200 words).

A brief sketch of where this work could lead in 5 years if this project succeeds and significant follow-on funding (upto $10M) were available. (< 150 words). Size of follow-up funding If your experiments through this award were to be successful, what scale of funding would you seek next to drive towards the aforementioned bigger ambition?

Please indicate (< 100 words) 3-5 measures that we should use to gauge the success of the project when it concludes. Project Narrative (pdf upload; up to 5 pages, excluding references) Please describe your proposed project in a pdf, and address the questions listed below. Proposals must not exceed 5 pages, including figures but excluding citations.

Problem and Opportunity : Which specific bottleneck are you solving, and for which application? Why does conventional compute underperform? Why is now the time to explore this approach?

(as opposed to 10 years ago, or 10 years from now). Approach and Methods : Describe the hardware being developed. What is the approach this project will take, and why will the approach lead to progress toward the challenge/problem?

Discuss why your project is appropriate for the budget track (Track I vs Track II) that you requested. Are there any risks associated with this approach, and how will you address this? Environmental Scan : How does your project fit in with the existing work in this field?

What are the new and distinguishing features of this approach? Existing Capability & Momentum: Describe your team and what you have already built or demonstrated. Include preliminary data, hardware, or results if available.

Workplan: Brief work plan with milestones. Budget: Provide a high-level budget breakdown. This is not a binding budget — detailed budgets will be requested from selected finalists.

We want to understand how you plan to allocate resources at a rough level. Is there anything else you'd like us to know? (optional) For example: links to demo videos, slides, relevant press coverage, prior collaborations with industry, suggestions or context that doesn't fit elsewhere (optional).

2026 Unconventional Compute RFP Request for Proposals: Unconventional Compute This is a pilot program. We are seeking evidence that unconventional compute hardware can tackle substantial, real-world problems that go beyond toy metrics. If this pilot uncovers compelling evidence, it may unlock a significantly larger investment in this space.

This is a lightweight application. There are no letters of support, no spreadsheets to be filled. The application consists of a short online form and a (less than) 5-page project narrative pdf.

We're interested in what you've built and where you want to take it—not your ability to write long proposals.

Schmidt Sciences seeks to fund research that can catalyze hardware fundamentally different from today’s CPU–GPU paradigm, and the training/inference methods co-designed to operate under the constraints imposed by such hardware: low numerical precision, device noise, drift, non-idealities etc. We are looking for proposals grounded in the creation of hardware that will solve specific, well-defined technological problems where conventional compute underperforms.

This request is open to universities and non-profits globally. You are building or designing compute hardware that is NOT based on conventional CPUs or GPUs. You have a real-world task in mind where your hardware could eventually outperform GPUs.

You are at a university or nonprofit research institution anywhere in the world. This RFP is not for you if: Your work is primarily about improving GPU-based architectures You don't yet have a target application in mind You are based at a for-profit institution.

April 30, 2026, 11:59PM Anywhere on Earth How to Apply Apply via SurveyMonkey Funding Tracks Track I: $50,000 - $150,000 ( 6 - 12 months) Track II: $150,000 - $750,000 (12 - 18 months) Estimated Number of Awards March 12, 2026, 10-11am EST. Register here March 19, 2026, 3-4pm EST. Register here April 7, 2026, 3-4pm EST.

Register here This request is open to universities and non-profits globally. aiforscience@schmidtsciences. org Frequently Asked Questions Indirect Costs Policy Indirect costs must not exceed 10% for the entire project.

Is there compute hardware that can outperform GPUs (latency, energy, throughput, etc.) in a technologically important niche? AI performance has been driven by scaling models and compute on CPU–GPU stacks, but the energy, latency, and cost of moving data is becoming the bottleneck. The next wave of computing need not look like a Transformer running on a GPU.

We believe that by exploiting novel physics and chemistry— for instance through optics, thermodynamics, or polymers—we can build hardware that solves specific, high-value problems that are intractable or inefficient with conventional compute.

This RFP seeks to identify and fund research that can catalyze hardware fundamentally different from today’s CPU–GPU paradigm, and the training/inference methods co-designed to operate under the constraints imposed by such hardware: low numerical precision, device noise, drift, and non-idealities.

We are looking for proposals grounded in the creation of hardware that will solve specific, well-defined technological problems where conventional compute underperforms. For example, some well-defined technological problems could include imaging flow cytometry, RF signal processing and high throughput industrial vision. We welcome applications across all domains.

We particularly encourage applications from teams with existing hardware, preliminary data, or demonstrated capabilities where additional resources would help drive towards more ambitious work. Successful projects that demonstrate valid physics and a credible demo will be considered for follow-on funding.

Funded teams will be invited to a convening of researchers and industry leaders in this space (anticipated Summer 2026), creating opportunities for collaboration and visibility. Applicants can apply for one of two funding track: Track I: $50,000 - $150,000 (6-12 months) Track I funds simulations, preliminary experiments, or proof-of-concept work that establishes the feasibility of a substantial prototype.

Successful Track I projects will produce evidence that a specific unconventional compute approach can address a real-world use case, with quantified performance projections.

Track I is appropriate if you: Have a credible approach but need to demonstrate feasibility before building hardware Need simulation or small-scale experimental data to validate performance claims Can produce preliminary evidence within 6-12 months Track II: $150,000 - $750,000 (12-18 months) Track II funds the creation of substantial prototypes that perform real-world functions.

Track II applicants must demonstrate that feasibility has already been established—through prior work, published results, or existing hardware. Track II is appropriate if you: Have existing hardware or strong preliminary data demonstrating feasibility. Are ready to build or scale a prototype for a specific application.

Can show meaningful progress within 12-18 months. We will fund projects focused on performance on a specific, substantial task(s).

In-scope approaches (illustrative, not exhaustive): Optical/photonic processing RF/microwave photonics and native RF-domain processing Resistive and mixed-signal compute Neuromorphic, stochastic/approximate computing In-scope task types (illustrative, not exhaustive): Real-time video / optical preprocessing and filtering at the sensor High-throughput microscopy classification/triage Always-on embedded sensing and event detection (industrial monitoring, wearables) Any other clearly defined application where a GPU is disadvantaged.

What we will not fund (generally) “Interesting hardware” without a compelling use case Toy use cases (e.g. MNIST digit recognition task) Proposals where the primary bottleneck is chemistry/unknown materials behavior Biology-based compute (e.g. DNA computing) Submissions will be assessed by internal and external reviewers with expertise relevant to your proposal topic.

During the assessment process, we may reach out to applicants to clarify portions of their proposal. Assessment criteria include: Problem Significance : Does the proposal identify a well-defined technological bottleneck (e.g., "ultra high speed industrial imaging") rather than a generic goal (e.g., "make AI faster")? Execution capability : team, facilities and proven ability.

What has the team already demonstrated? Is there existing momentum? Well-characterized physical phenomena and materials : We deprioritize proposals whose success depends heavily on the discovery of novel materials or chemistries.

Scalability : What is the likelihood that the proposed research, in the long term, will create new technology? Documents should be submitted through the SurveyMonkey Apply portal. Below is a list of the items you will need to submit: Name, email and title of the person submitting the proposal.

Institution/Recipient of Fund. Choice of Funding Track (I or II) Total Project Budget (USD) Project Start Date (MM/YY) Please upload the CV of the lead PI (mandatory) and other team members (optional) (Must be uploaded as a PDF.) Please provide a short abstract of your project written for a non-specialist audience (< 200 words).

A brief sketch of where this work could lead in 5 years if this project succeeds and significant follow-on funding (upto $10M) were available. (< 150 words). Size of follow-up funding If your experiments through this award were to be successful, what scale of funding would you seek next to drive towards the aforementioned bigger ambition?

Please indicate (< 100 words) 3-5 measures that we should use to gauge the success of the project when it concludes. Project Narrative (pdf upload; up to 5 pages, excluding references) Please describe your proposed project in a pdf, and address the questions listed below. Proposals must not exceed 5 pages, including figures but excluding citations.

Problem and Opportunity : Which specific bottleneck are you solving, and for which application? Why does conventional compute underperform? Why is now the time to explore this approach?

(as opposed to 10 years ago, or 10 years from now). Approach and Methods : Describe the hardware being developed. What is the approach this project will take, and why will the approach lead to progress toward the challenge/problem?

Discuss why your project is appropriate for the budget track (Track I vs Track II) that you requested. Are there any risks associated with this approach, and how will you address this? Environmental Scan : How does your project fit in with the existing work in this field?

What are the new and distinguishing features of this approach? Existing Capability & Momentum: Describe your team and what you have already built or demonstrated. Include preliminary data, hardware, or results if available.

Workplan: Brief work plan with milestones. Budget: Provide a high-level budget breakdown. This is not a binding budget — detailed budgets will be requested from selected finalists.

We want to understand how you plan to allocate resources at a rough level. Is there anything else you'd like us to know? (optional) For example: links to demo videos, slides, relevant press coverage, prior collaborations with industry, suggestions or context that doesn't fit elsewhere (optional).

Feb 13 2026 12:00 AM (EST) Apr 30 2026 11:59 PM (EDT)