Cactus Compute
Machine Learning · AI Compute & Inference
Overview
On-device AI with cloud fallback. Cactus post-trains models to know when they are wrong and hand off to frontier cloud models — cutting inference costs up to 5x.
Funding and backers
Backers
1
Y Combinator
Momentum
15/100
rising traction
Last sign of life
2 mo ago
a posting or a news item
The story so fareach stage dated by its source: the company's releases, the press, registers, filings
Oct 8
On the Hacker News front page Whistle: Speech to Text in 16.9 MB · 964 points
Aug 10
Launched on Hacker News Show HN: Needle2: 14MB agentic LLM for phones, wearables, smart home and robots
Jun 2025
Went through Y Combinator
Sep 2024
Website went live first capture in the Wayback Machine
Tractionpublic signals, read monthly
Registers and programsofficial lists and records that name the company
| Legal entity | Cactus Compute, Inc. | its site |
Backed byfrom the funds' portfolios and accelerator catalogs
Investors in similar startupsfunds that back the companies most like this one
Similar startupsby what they do, their industry and backers
ENERZAi is a South Korean AI software company that optimises AI models for on-device inference, delivering low-memory edge AI agents…
webAI is an artificial intelligence platform company headquartered in Austin, Texas, and founded in 2019 by David Stout, Tyler Mauer, and…
Base Compute is an AI inference lab. We build the runtimes and infrastructure that make powerful AI run on-device. Fast, private, and at…
Deploy on-device AI to deliver fast, adaptive, and private user experiences — at global scale.
Cephable is an on-device AI assistant that controls, creates, and automates your work. Runs locally on your hardware. No cloud, nothing…
OpenInfer is an innovation lab building one unified inference stack, from kernel to cloud — reimagining inference around the economics of…
Connections
List
Graph
Found in company registers, site code, DNS and hiring. Each connection shows how sure we are and the evidence, with its source and date.
Shared investors
Y CombinatorVia a fund
6310 portfolio companies
Click a company to open it. Dashed lines are likely connections. The second step of the network is part of the Pro plan.
Owner
Same company
Same group
Likely
News
Needle 2 - The 14 MB Agentic LLM for Tiny Devices Cactus
An open 45M-parameter model for tool calling, device use, and structured extraction. Needle 2 runs as a 14 MB binary in 28 MB of session RAM.
Read more
Report
GitHub - cactus-compute/cactus: Quantization, kernels, runtime and inference engine for mobiles, wearables, smart home and robots.
Quantization, kernels, runtime and inference engine for mobiles, wearables, smart home and robots. - GitHub - cactus-compute/cactus: Quantization, kernels, runtime and inference engine for mobiles, wearables, smart home and robots.
Read more
Report
Offices
Where the company hires and what each office is for
Cities
1
Hiring now
0
Countries
1
San Francisco
United States
HQ