What it sells
Sparse-native compute for AI: processors, boards, software and multi-rack systems that compute only what matters, so sparse models run faster and cheaper per token. Claims 30-44x faster than incumbent hardware on sparse models.
Business model breakdown
Distribution: how it reaches customers
Partner with us (direct partnership/enterprise sales); systems deployed to customers.
Funding history and coverage
| Round date | Stage | Amount | Investors | Source |
|---|---|---|---|---|
| 2026-11-11 | Series A | $90M total | — | news.google.com |
| 2026-11-11 | Series A | $90M | — | news.google.com |
| 2026-11-11 | Series A | $90M total | — | news.google.com |
| 2026-11-10 | Series A | $90M total | — | news.google.com |
| 2026-11-10 | Series A | $90M total | — | news.google.com |
| 2026-11-10 | Series A | $90M total | — | news.google.com |
| 2026-11-10 | Series A | $90M total | — | news.google.com |
| 2026-11-10 | — | €80M | — | news.google.com |
| 2026-11-10 | Series A | $90M total | — | news.google.com |
| 2026-11-10 | Series A | $90M total | — | news.google.com |
| 2026-11-10 | Series A | $90M total | — | news.google.com |
| 2026-11-10 | Series A | $90M total | — | news.google.com |
| 2026-11-10 | — | €80M | — | eu-startups.com |
| 2026-11-10 | — | — | — | news.google.com |
| 2026-11-10 | — | $90M total | Lockheed Martin Ventures | techfundingnews.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 2026-11-12.