Infrastructure / Deployment

Daytona Sandbox

Open-source agent-native compute platform that provisions sub-200ms stateful sandboxes (start, pause, fork, snapshot, destroy) for AI agents. Picked over E2B for snapshot/fork semantics and an enterprise BYOC story. Raised $24M Series A in Feb 2026 led by FirstMark with strategic investment from Datadog and Figma Ventures, ~72k stars on GitHub.

Daytona secure/elastic runtime for AI-generated code execution; v0.184.0 (Jun 3 2026) live on GitHub June 2026. Repo now AGPL-3.0 (note: Daytona pivoted from dev-environments to agent code-execution sandboxes). Sub-90ms sandbox spin-up; managed cloud at app.daytona.io.

Openness

4 high confidence

4.0

license: AGPL-3.0(OSI, copyleft)
source: public
managed-tier: app.daytona.io hosted cloud + hybrid (proprietary)
core-gated: gated

AGPL-3.0 (OSI-approved copyleft) core plus a proprietary managed cloud => open_core. Scored 4 alongside the Apache-licensed open-core peers (E2B, Ray, Coder): OSI copyleft is not treated as less open on this map, since a copyleft obligation binds redistributors rather than capping who may use the software or at what scale. Corrected from 3, which had penalized copyleft for 'constraining commercial reuse' - a rationale the other 22 copyleft software products do not apply.

https://github.com/daytonaio/daytona recorded 2026-06-04
AGPL-3.0 license; secure runtime for AI-generated code; managed cloud at app.daytona.io

Adoption

4 high confidence

4.0

PyPI: ~8.31M downloads last month for the daytona package (primary). Note part of this volume is inherited from the prior dev-environment product line, but the package now ships the sandbox SDK. Solidly 1M-10M band.

https://pypistats.org/packages/daytona recorded 2026-06-04
~8,313,410 downloads last month

Capability

4 high confidence

4.0

Strong isolation (dedicated kernel + isolated net/fs, allocated vCPU/RAM/disk) with sub-90ms cold-start and persistent sessions; among the strongest agent-runtime feature sets, just below the kernel/microVM frontier-definers. Independent ComputeSDK TTI benchmarks corroborate the sub-100ms-class cold-start (~140ms median, 4th-fastest of 19 sandbox providers).

https://github.com/daytonaio/daytona recorded 2026-06-04
dedicated kernel/filesystem/network stack per sandbox; <90ms spin-up; persistent state; Python/TS/JS
https://www.computesdk.com/benchmarks/sandboxes/ recorded 2026-07-09
ComputeSDK TTI leaderboard (independent, 100 iters/day, 2026-07-09 run): median time-to-interactive ~140ms sequential / ~710ms burst @100 concurrent; 4th-fastest of 19 providers

Unchanged since 2026-07-30 (last edited, not re-checked)