NVIDIA Inception is NVIDIA's free accelerator program for startups building on AI, data science, and high-performance computing. Pearl Research Labs is a member of this program.
You'll see the Inception badge on our website, so it's worth explaining what it does and — just as importantly — what it doesn't.
What the program is
Inception is a virtual, always-open program rather than a cohort-based accelerator. There are no application deadlines, no batches, and no demo day. As of 2026 it spans more than 40,000 member companies, making it one of the largest startup networks in technology.
What members get:
- Preferred pricing on NVIDIA hardware and software
- Technical resources — early access to developer tools and SDKs
- Deep Learning Institute credits — free self-paced courses and discounted instructor-led workshops
- Partner offers, including access to programmes like AWS Activate with cloud credits
- Exposure to NVIDIA's investor network
⚠️ What Inception membership is not
We want to be direct about this, because program badges get over-read — sometimes by projects that benefit from the confusion.
- It is not an investment. NVIDIA takes no equity, and no money changes hands in either direction. The program is explicitly free with no strings attached.
- It is not a partnership in the commercial sense, and it doesn't imply NVIDIA is building with us or on us.
Why we joined anyway
Because it's genuinely useful for what we build, and free.
Pearl is unusually GPU-dependent even by AI-startup standards. Our mining kernel operates inside the matrix multiplications of a model's forward pass, which means we work at the level of CUDA, GPU memory behaviour, and accelerator architecture. Our published benchmarks run on H200s. Practical access to NVIDIA's developer resources and hardware pricing matters more to us than to a typical software company.
The wider network has value too — Pearl only works if inference providers adopt it, and Inception's membership is largely companies running GPU fleets. That's our audience.