What it sells
Fastest inference anywhere: open models on-prem, hosted or on-device
Business model breakdown
Buyer: who owns the budget?
UnverifiedDevelopers and engineering leads building AI apps who need fast or private inference.
Trigger: what makes them buy now?
Cloud AI is too slow, expensive or fragile for real products and they want on-device or controlled hosted execution.
Workflow: what inputs become what output?
Developer copies the SKILL.md into their agent or uses the SDK, signs in to the console, buys credits, then sends OpenAI-compatible requests to hosted Wally models or runs models locally on Apple Neural Engine / Qualcomm NPU through open-source SDKs; usage and spend are tracked in the console.
Economics: what it saves and how it charges
Prepaid credits for hosted inference (Wally) bought in the console; the SDKs and local runtimes are open source, which drives adoption of the paid hosted tier.
Distribution: how it reaches customers
Self-serve: console sign-in, buy credits, run; open-source SDKs (Swift, Kotlin, React Native, Flutter, TypeScript, C++) on GitHub plus agent-readable SKILL.md; Talk to us for sales.
Funding history and coverage
| Round date | Stage | Amount | Investors | Source |
|---|---|---|---|---|
| — | Y Combinator Winter 2026 | — | Y Combinator | ycombinator.com |
Profiles summarize public reporting and company websites. Business-model notes are MaxBC's reading of that material, not confirmed facts; unknowns are left blank rather than guessed. Updated 2026-10-12.