JulesvsSWE-agent
Generated from the two spec rows. Green marks the better value where a spec has a clear direction. Everything else is just different.
Jules
Google · Autonomous SWE
#141MCP
Panel
5.72 spec wins
- Reliability
- 5.3
- Usefulness
- 5.8
- Cost
- 6.3
- Longevity
- 5.3
“Fifteen free tasks a day, as long as you are over 18 and using a personal Gmail rather than the company account.”
SWE-agent
Princeton University and Stanford University · Autonomous SWE
#144OSS
Panel
5.15 spec wins
- Reliability
- 5.5
- Usefulness
- 4.8
- Cost
- 6.5
- Longevity
- 3.7
“Maintenance-only and superseded by a version with mini in the name, which is the most honest changelog on this board.”
Spec by spec
| Spec | Jules | SWE-agent |
|---|---|---|
| Architecture | ||
| Category | Autonomous SWE | Autonomous SWE |
| Runs | cloud, sandbox | local, sandbox |
| Platforms | web, macos, linux, windows | macos, linux |
| Context window | not documented | not documented |
| Protocols | ||
| MCP client | Yes | No |
| MCP server | No | No |
| Capabilities | ||
| Runs terminal commands | Yes | Yes |
| Multi-file edits | Yes | Yes |
| Git operations | Yes | Yes |
| Browser control | YesThe default VM image includes Playwright and chromedriver, and Jules renders a front end and returns screenshots to verify its work (https://jules.google/docs/environment). | No |
| Sandboxed execution | YesEach task gets a secure, short-lived Ubuntu VM with internet access (https://jules.google/docs/environment). | Yes |
| Multi-agent orchestration | No | No |
| Headless / CI mode | YesScriptable through `jules remote new --repo ... --session "..."` and the REST API (https://jules.google/docs/cli/reference). | Yes |
| Models | ||
| Backbone | Gemini 3 Flash, Gemini 3.1 Pro | Claude, GPT, any LiteLLM-supported model |
| Bring your own model | NoJules runs only on Google-hosted Gemini models; there is no provider picker, custom base URL, gateway or cloud deployment of your own. | Yes |
| Local models | NoTasks run in a Google-hosted VM with no configurable model endpoint, so no local LLM can be used. | Yes |
| Cost | ||
| Pricing model | subscription | byok |
| Starts at | $19.99/mo | $0/mo |
| Free tier | Yes | Yes |
| Bring your own key | NoAccess comes from a Google AI plan; no Gemini API key or Google Cloud project of your own can be attached. | Yes |
| Openness | ||
| Open source | No | Yes |
| License | proprietary | MIT |
| GitHub stars | n/a | 20,455 |
Which one would each critic pick
| Critic | Jules | SWE-agent | Pick |
|---|---|---|---|
| El Juez | — | — | not enough reviews |
| El Amigo | 6.8 | 5.0 | Jules — Use Jules for the backlog you never get to, fifteen free tasks a day on a personal Google account, and do not adopt it as a team tool because the paid plans do not accept business accounts yet. |
| El Crítico | 6.3 | 4.8 | Jules — Jules runs your repo in a VM you cannot inspect on a model you cannot choose, and the failure mode is a pull request built on an environment that never installed correctly. |
| El Profesor | 6.3 | 5.8 | Jules — Jules has the correct autonomous architecture, an isolated VM, a plan reviewed before edits, tests run in place, and publishes no benchmark by which to judge what it does with it. |
| La Inversora | 6.0 | 4.0 | Jules — Jules is a feature of a consumer subscription, not a company, and the risk to a buyer is a deprecation notice rather than a bankruptcy. |
| La Jefa | 4.8 | 3.8 | Jules — Paid plans are limited to individual Google accounts and Workspace support is in development, so there is no way to buy this for sixty engineers; not yet. |
| El Hacker | 4.3 | 7.5 | SWE-agent — MIT, Docker-sandboxed, any LiteLLM model including my local server, and every prompt in the repo; maintenance-only just means the code stops changing under me. |
Picks are derived from each critic's own scores. Humans vote on matchups on the duels page.