StartupSan Francisco, CA · Series A
Simular
Computer-use agents: the open-source Agent S framework, the Sai agent and the Simular API.
Simular, founded by former DeepMind scientists, builds computer-use agents that operate desktop and web apps by looking at the screen and clicking, typing and scrolling. Its open-source Agent S framework runs on macOS, Windows and Linux; the project reports Agent S3 as the first to surpass human performance on OSWorld (72.60%) and released Agent S4 in October 2026. Commercially it offers Sai, a production computer-use agent, and the Simular API for giving agents autonomous cloud computers. In December 2025 it announced a $21.5M Series A led by Felicis, with NVentures and South Park Commons participating.
Pros & cons
Each point cites its sources and says what kind of evidence backs it. A second reviewer tried to refute every claim before it was kept.
Pros
- MetadataIts Agent S framework is open source (Apache-2.0, about 12.8K stars) and runs on macOS, Windows and Linux.[1]
- DocsAgent S reported about 72.6% on OSWorld, around the human baseline, and the Agent S2 work is published on arXiv.[1][2]
- Independent testSai operates full desktops, browsers and terminals, not just web pages.[1]
Cons
- Independent testSai's 73% OSWorld 2.0 result was first published on Simular's own site and was not yet on the official leaderboard.[1]
- DocsSimular says pure neural agents are expensive, non-deterministic and slow, and that variance between runs is the core bottleneck to usefulness.[1]
- User reportsSai signups were limited to control costs, and the regular price was quoted at $200/month.[1]
- Independent testThird-party reviews are mixed, with one scoring Sai 5.5/10 for notable limitations.[1]
- Best for
- Researchers and teams automating desktop GUI workflows who want an open-source computer-use framework with a commercial option.
- Watch out for
- Reliability gap between benchmark scores and repeated real tasks, plus the cost of the hosted agent.
Assessed Oct 29, 2026 · adversarially reviewedSuggest a correction
People
Co-founder & CEO
Harness engineering
Co-founder
Harness engineering
In research (1)
- Evolving Agent InterfacesVerdict: Useful
Oct 20, 2026 · Authors' company; Agent S4 extends its Agent S line
Neighbours in the directory
San Francisco, CA · Est. 2023
Conversational AI agents for customer experience.
Sierra, co-founded by Bret Taylor and Clay Bavor, builds production conversational agents that resolve real customer issues with guardrails, evaluation and human escalation — a template for deploying vertical agents at the enterprise edge.
~$10B valuation (Oct 2025)
San Francisco, CA · Est. 2023
The integration layer: hundreds of managed tools your agents can call.
Composio handles auth, schemas, observability and reliability for the external SaaS tools an agent uses — 'plumbing for agentic actions' across Slack, GitHub, Salesforce and 250+ other surfaces.
Est. 2024
A registry and toolchain for MCP servers.
Smithery indexes and helps deploy MCP (Model Context Protocol) servers — a distribution layer for the tool ecosystem agents connect to.
Sources
Verified Oct 29, 2026