Cumulus Labs
YC W26 · Active · The Fastest Multimodal Inference OS
Competitors
| Company | Category | Status | Program |
|---|---|---|---|
 Airtrain AIShared: fine tuning, evaluation, productionInactiveAI, Devtools | AI, Devtools | Inactive | YCS22 |
 TrainLoopShared: fine tuningActiveAI, Devtools | AI, Devtools | Active | YCW25 |
 Tagflow AIShared: fine tuningNot statedAI | AI | Not stated | Antler2024 |
 CTGTShared: fine tuning, fail, engineeringActiveAI, Devtools | AI, Devtools | Active | YCF24 |
 PolymorphShared: fine tuning, inference, engineeringActiveAI | AI | Active | YCW26 |
 Together AInot in a program hereShared: fine tuningElsewhereOther | Other | Elsewhere | |
 Positronnot in a program hereShared: inferenceElsewhereOther | Other | Elsewhere |
Adjacent
| Company | Category | Status | Program |
|---|---|---|---|
 SazabiShared: observability, engineering, infrastructureActiveAI, Devtools | AI, Devtools | Active | YCX26 |
 OneTriangleShared: inference, fastest, infrastructureActiveAI, Devtools | AI, Devtools | Active | YCS26 |
 Zibra LabsShared: fine tuning, inference, buildingActiveAI, Devtools | AI, Devtools | Active | YCX26 |
 ReflexShared: needing, fragmented, unifiedActiveAI, Devtools | AI, Devtools | Active | YCW23 |
 HopperShared: inference, production, infrastructureActiveAI, Devtools | AI, Devtools | Active | YCF26 |
Also from W26 in AI
All of W26| Company | Category | Status |
|---|---|---|
 /dev/fast/dev/fast is the AI-native code forgeActiveAI, Devtools | AI, Devtools | Active |
 10x ScienceThe AI-native platform for next-generation protein characterization.ActiveAI | AI | Active |
 21stUI building blocks for the agentic internetActiveAI, Devtools | AI, Devtools | Active |
 AemonThe Forward-Deployed AI Research EngineerActiveAI, Devtools | AI, Devtools | Active |
 Agentic FabriqThe control plane for AI agents.ActiveAI | AI | Active |
 Alt-XBuilding the best venue for private markets exposure.ActiveAI, Data | AI, Data | Active |
Founders
- Veer ShahFounder/CEO
About Cumulus Labs
Cumulus Labs lets engineering teams ship AI in production without needing a dedicated ML platform team. Right now, companies building AI products are forced to stitch together separate vendors for routing, observability, evaluation, fine-tuning, and inference. This fragmented approach is brittle, expensive, and is a common reason enterprises fail with AI. We replace that entire stack with a single unified platform. Developers can keep their existing code while instantly upgrading to a unified platform that handles routing, semantic caching, continuous shadow evaluation, simulated data, and one-click fine-tuning. Behind the platform is Ion, our proprietary inference engine running on a custom NVIDIA Grace GPU fleet. Ion uses in-house custom GPU kernels to deliver 30 to 50 percent more throughput than standard vLLM or SGLang, giving our customers SOTA inference economics.
- YC batch
- W26 · Winter 2026
- Status, per YC
- Active
- Category
- AI, Developer tools
- Team size
- 2
- Location
- San Francisco, CA, USA
- Founded
- 2025
- Checked
- 1h ago
Cumulus Labs: questions
Which YC batch was Cumulus Labs in?
Cumulus Labs was in Y Combinator's Winter 2026 batch (W26).
Is Cumulus Labs still active?
Y Combinator lists Cumulus Labs as active.
Who founded Cumulus Labs?
Cumulus Labs's founders, as listed on its YC page: Veer Shah (Founder/CEO).
Where is Cumulus Labs based?
Cumulus Labs lists San Francisco, CA, USA as its location on YC's directory.
Sources: each program's public company list, read twice a week. Not affiliated with Y Combinator or any other program. Logos via logo.dev; trademarks belong to their owners.