AidervsPoolside
Generated from the two spec rows. Green marks the better value where a spec has a clear direction. Everything else is just different.
Aider
Aider AI (Paul Gauthier) · Terminal agent
#7OSS
Panel
7.12 spec wins
- Reliability
- 7.2
- Usefulness
- 6.8
- Cost
- 7.5
- Longevity
- 7.0
“Commits every change to git so you can see exactly which one broke the build.”
Poolside
poolside · Terminal agent
#15MCP
Panel
6.93 spec wins
- Reliability
- 6.8
- Usefulness
- 7.0
- Cost
- 6.5
- Longevity
- 7.2
“One of its three network policies is named unsafe-allow-all, which is at least the most clearly labelled decision on this board.”
Spec by spec
| Spec | Aider | Poolside |
|---|---|---|
| Architecture | ||
| Category | Terminal agent | Terminal agent |
| Runs | local | local, sandbox, cloud |
| Platforms | macos, linux, windows | macos, linux, windows |
| Context window | not documented | 1M tokens on Laguna S 2.1, 256k on XS 2.1 and M.1 |
| Protocols | ||
| MCP client | No | Yes |
| MCP server | No | No |
| Capabilities | ||
| Runs terminal commands | Yes | Yes |
| Multi-file edits | Yes | Yes |
| Git operations | Yes | Yes |
| Browser control | YesThe /web command and URL detection drive a real Chromium through Playwright to render and read pages (--disable-playwright turns it off); the --browser flag is only aider's own Streamlit GUI. | NoWeb search and fetch are documented; there is no DOM-level browser automation. |
| Sandboxed execution | NoAider publishes a Docker image as an install option only; commands it runs are not sandboxed. | YesTool commands run inside a container on your machine and need a local Docker engine; network policy is off, allow-list or unsafe-allow-all, enforced through a proxy container. |
| Multi-agent orchestration | No | Yes |
| Headless / CI mode | Yes | Yes`pool exec --prompt ... --output json` with POOLSIDE_API_KEY, plus a documented GitHub Actions integration. |
| Models | ||
| Backbone | OpenAI, Anthropic, Gemini, DeepSeek, OpenRouter, Bedrock, Vertex AI, Azure, Groq, xAI, Ollama, LM Studio | Laguna S 2.1, Laguna XS 2.1, Laguna M.1 |
| Bring your own model | Yes | Yes |
| Local models | Yes | YesDocumented local-run guides for Ollama and vLLM on Metal, plus full on-premises and air-gapped deployment. |
| Cost | ||
| Pricing model | byok | mixed |
| Starts at | $0/mo | n/a |
| Free tier | Yes | Yes |
| Bring your own key | Yes | Yes |
| Openness | ||
| Open source | Yes | No |
| License | Apache-2.0 | proprietary |
| GitHub stars | 49,324 | n/a |
Which one would each critic pick
| Critic | Aider | Poolside | Pick |
|---|---|---|---|
| El Juez | — | — | not enough reviews |
| El Amigo | 7.5 | 6.8 | Aider — Pick Aider if you want to see every diff and steer every step; pick Claude Code if you want the agent to find the files and run on its own. |
| El Crítico | 7.3 | 6.5 | Aider — Every edit is a real edit on your disk with no sandbox, and auto-commit is the one mitigation standing between a bad command and your afternoon. |
| El Profesor | 8.3 | 6.5 | Aider — The most legible architecture on the board, with a published harness; the SWE-bench numbers are from 2024 and should be read as history, not as a claim. |
| La Inversora | 5.3 | 7.0 | Poolside — Training your own models and selling deployments into other people's data centres is a capital-intensive bet on the one segment that cannot buy from a lab directly. |
| La Jefa | 5.5 | 7.0 | Poolside — It runs inside our own network, including air-gapped, on the Kubernetes platforms we already operate, and the enterprise price is a sales conversation. |
| El Hacker | 9.0 | 7.5 | Aider — Python I can read, prompts in the repo, .aider.conf.yml for model and edit format, --model ollama/anything for an air-gapped box; MCP is the one gap. |
Picks are derived from each critic's own scores. Humans vote on matchups on the duels page.