PoolsidevsWarp
Generated from the two spec rows. Green marks the better value where a spec has a clear direction. Everything else is just different.
Poolside
poolside · Terminal agent
#15MCP
Panel
6.90 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.”
Warp
Warp · Terminal agent
#20OSSMCP
Panel
6.52 spec wins
- Reliability
- 6.5
- Usefulness
- 6.7
- Cost
- 5.5
- Longevity
- 7.2
“Scored 75.8 on SWE-bench Verified with a best-of-k wrapper, which is also how most of us get through code review.”
Spec by spec
| Spec | Poolside | Warp |
|---|---|---|
| Architecture | ||
| Category | Terminal agent | Terminal agent |
| Runs | local, sandbox, cloud | local, cloud |
| Platforms | macos, linux, windows | macos, linux, windows |
| Context window | 1M tokens on Laguna S 2.1, 256k on XS 2.1 and M.1 | 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 | NoWeb search and fetch are documented; there is no DOM-level browser automation. | YesComputer Use environments bundle Chromium plus the Playwright CLI and can attach over CDP, so the agent navigates, clicks, fills forms and reads pages. |
| Sandboxed execution | 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. | YesCloud runs execute in a containerized sandbox built from a Docker image with no access to the local machine; local interactive sessions are not sandboxed. |
| Multi-agent orchestration | Yes | Yes |
| Headless / CI mode | Yes`pool exec --prompt ... --output json` with POOLSIDE_API_KEY, plus a documented GitHub Actions integration. | Yes |
| Models | ||
| Backbone | Laguna S 2.1, Laguna XS 2.1, Laguna M.1 | GPT, Claude, Gemini, Grok, open-weight models via Fireworks |
| Bring your own model | Yes | YesAny OpenAI-compatible endpoint (OpenRouter, LiteLLM, an internal gateway) plus BYO Anthropic, OpenAI or Google keys and custom routers. |
| Local models | YesDocumented local-run guides for Ollama and vLLM on Metal, plus full on-premises and air-gapped deployment. | YesOllama, LM Studio, vLLM and llama.cpp are supported through the custom inference endpoint, but only when exposed at a public URL via a tunnel, since localhost and private addresses are rejected. |
| Cost | ||
| Pricing model | mixed | mixed |
| Starts at | n/a | $20/mo |
| Free tier | Yes | Yes |
| Bring your own key | Yes | Yes |
| Openness | ||
| Open source | No | Yes |
| License | proprietary | AGPL-3.0 |
| GitHub stars | n/a | 65,338 |
Which one would each critic pick
| Critic | Poolside | Warp | Pick |
|---|---|---|---|
| El Juez | — | — | not enough reviews |
| El Amigo | 6.8 | 7.3 | Warp — Use Warp if you want an agent in the terminal you already open every morning, with a standalone CLI and cloud agents when you outgrow it, and read the credit table before you pick a plan. |
| El Crítico | 6.5 | 6.0 | Poolside — Isolation needs a container engine running on the machine and enforces network policy through a proxy container, which is a configuration rather than a boundary. |
| El Profesor | 6.5 | 6.0 | Poolside — The vendor trains its own model family and publishes no evaluation of it anywhere on the row, which is a conspicuous silence for a company selling its own inference. |
| La Inversora | 7.0 | 6.8 | 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 | 7.0 | 6.0 | 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 | 7.5 | 6.8 | Poolside — The client is proprietary and the model weights are published under OpenMDW-1.1 and Apache-2.0, which is the exact opposite of everyone else on this board. |
Picks are derived from each critic's own scores. Humans vote on matchups on the duels page.