Wafer Raises $40M Series A, Hits $8M ARR in Four Months

YC-backed AI inference cloud Wafer announced a $40M Series A led by Marathon and Chemistry, with Y Combinator participating, at a valuation above $200M; according to @stephpalazzolo, the company had earlier declined acquisition offers. Its cloud service reached $8M ARR from zero within four months of launch, making it one of the fastest-growing newcomers in inference.

Confirmed

• The $40M round was co-led by Marathon and Chemistry with YC participating, at a valuation over $200M (per @stephpalazzolo, @ycombinator)
• ARR grew from zero to $8M within four months of launch (per @ycombinator, @garrytan)
• Founded about a year ago; previously rejected acquisition offers (per @stephpalazzolo)
• Its approach is "AI optimizing AI": agents continuously tune GPU clusters to run open-source models at industry-leading speed, combining open-source cost with small-model latency; uses both Nvidia and non-Nvidia chips (per @ycombinator, @stephpalazzolo)
• The company began as a university side project, discussed on the YC founder interview show (per @ycombinator)

Why it matters

• Inference cost and latency are key bottlenecks for open-source model adoption at scale; if Wafer's agent-driven GPU tuning proves out, it could reshape inference cloud economics
• The $8M ARR in four months signals strong downstream demand, with the inference layer becoming a new capital magnet

2026-09-01 ~ 2026-09-02 · 6 related posts

• [source] Inference provider Wafer raises $40m at $200m+ valuation, rejecting offers — steph_palazzolo · 2026-09-01
• Inference optimization startup Wafer AI raises $40M Series A — ycombinator · 2026-09-02
• [source] Wafer raises $40M Series A, hits $8M ARR optimizing AI inference — ycombinator · 2026-09-02

3 near-duplicate retellings: ycombinator · ycombinator · garrytan