kimchivsOpenWorker
Generated from the two spec rows. Green marks the better value where a spec has a clear direction. Everything else is just different.
kimchi
kimchi · Terminal agent
#75OSSMCP
Panel
6.21 spec wins
- Reliability
- 6.2
- Usefulness
- 6.8
- Cost
- 5.8
- Longevity
- 5.8
“It divides work between an orchestrator, a builder and an explorer, which is more organisational structure than some of its users have.”
OpenWorker
Andrew Ng · Terminal agent
#121OSSMCP
Panel
6.23 spec wins
- Reliability
- 5.3
- Usefulness
- 6.3
- Cost
- 6.8
- Longevity
- 6.3
“A local-first agent with full access to your desktop, files, and terminal that runs without a sandbox.”
Spec by spec
| Spec | kimchi | OpenWorker |
|---|---|---|
| Architecture | ||
| Category | Terminal agent | Terminal agent |
| Runs | local | local |
| Platforms | macos, linux, windows | macos, windows, linux |
| Context window | not documented | not documented |
| Protocols | ||
| MCP client | Yes | Yes |
| MCP server | No | No |
| Capabilities | ||
| Runs terminal commands | Yes | Yes |
| Multi-file edits | Yes | Yes |
| Git operations | Yes | Yes |
| Browser control | No | No |
| Sandboxed execution | No | No |
| Multi-agent orchestration | YesMulti-model mode assigns separate models to orchestrator, builder and explorer roles and delegates tasks between them. | Yes |
| Headless / CI mode | YesThe repository ships a Terminal-Bench adapter that aggregates usage across session files, which implies unattended runs. | No |
| Models | ||
| Backbone | kimchi-dev models, Anthropic | OpenAI, Anthropic, Google Gemini, BytePlus Ark, Volcengine Ark Agent Plan, Inkling, GLM, DeepSeek |
| Bring your own model | Yes | Yes |
| Local models | No | Yes |
| Cost | ||
| Pricing model | usage | free |
| Starts at | n/a | n/a |
| Free tier | No | Yes |
| Bring your own key | Yes | Yes |
| Openness | ||
| Open source | Yes | Yes |
| License | Apache-2.0 | MIT |
| GitHub stars | 2,231 | 18,394 |
Which one would each critic pick
| Critic | kimchi | OpenWorker | Pick |
|---|---|---|---|
| El Juez | — | — | not enough reviews |
| El Amigo | 6.5 | 7.5 | OpenWorker — Pick OpenWorker for its local-first security agents and broad model support if you accept the risks of running an unsandboxed AI on your desktop. |
| El Crítico | 6.0 | 5.5 | kimchi — In multi-model mode an orchestrator delegates to role-specific models, so a bad result has three possible authors and nothing documented attributes it. |
| El Profesor | 6.5 | 6.5 | no preference |
| La Inversora | 6.0 | 5.8 | kimchi — 2,220 stars, a permissive CLI and an API key in the middle: the code is the marketing and the inference is the company. |
| La Jefa | 5.0 | 4.0 | kimchi — No published prices anywhere and usage-based billing behind an API key, so I cannot model sixty engineers, which ends the conversation before security starts. |
| El Hacker | 7.0 | 8.0 | OpenWorker — OpenWorker is a proper desktop agent harness: MIT-licensed, local-first, BYOM, and you can run it from source. But its power is a risk without a sandbox. |
Picks are derived from each critic's own scores. Humans vote on matchups on the duels page.