Gimlet raises $300M at $3B valuation to end idle compute in AI inference
AccBalanced · x · 2026-09-05
Inference startup Gimlet raised a $300M Series B led by a16z at a $3B valuation, with Samsung and Hudson River Trading participating — the latter's HFT business depends heavily on inference latency.
The company's two founding bets:
- AI inference will become the majority workload in software overall, not just in AI
- Inference is multiple orders of magnitude less efficient than it should be
Gimlet's approach: decompose models and schedule the parts across different hardware so no compute sits idle. Founder @theojaffee: "You don't want any idle compute. Compute is so valuable."
More from Venture
- Vanta CEO on the 'boring and complex' playbook, eating AI costs, and 10X comp — soleio · 2026-10-21
- VC investor argues massive funding rounds reflect over-financialized private markets — arian_ghashghai · 2026-10-21
- AI SaaS monetization is cooked: consumers small, enterprises pay $20/month — menhguin · 2026-10-21
- Stanford professor releases 2026 AI Venture Ranking of top 50 VC firms — mmurph · 2026-10-21
- The AI revenue death-spiral scenario: capex, rate pressure, and startups cutting token spend — menhguin · 2026-10-21
- AI bubble author clarifies: legit ROI use cases, but heavy users are all inside the AI ecosystem — menhguin · 2026-10-21