# poolside.ai

> AI-optimized mirror of poolside.ai containing 155 pages totalling 282,640 words of clean markdown content, structured data, and semantic HTML. Original source: https://poolside.ai. Last updated: 2026-07-26T06:15:53.593Z.

Each page is available as HTML (with JSON-LD structured data) and Markdown (text-only, ideal for LLMs and RAG).

## Homepage

- [Poolside Platform](/content/platform/index.html) (23 words)
- [Poolside Chat](/content/chat/index.html): Move from ideas to answers with Poolside. Chat with Laguna AI models, explore topics, analyze files, and work through complex information. (14 words)
- [Poolside Trust Center | Powered by Drata](/content/trust/index.html): Ready to turn trust into your competitive advantage? Sprint through security reviews and quickly share key security information with Trust Center. (133 words)
- [Trajectories — Poolside](/content/trajectories/index.html): Explore trajectories from the evaluation of Poolside models as they are released. (879 words)
- [Poolside Support](/content/support/index.html) (38 words)
- [Poolside Documentation - Poolside](/content/docs/index.html): Build with Poolside agentic coding models in your terminal, editor, tools, or applications. (88 words)
- [Poolside](/content/site-root.html): Poolside is a foundation model company bringing intelligence to everywhere work gets done. Our mission is to drive abundance for humanity by creating artificial general intelligence., Poolside is a foundation model company bringing intelligence to everywhere work gets done. Our mission is to drive abundance for humanity by creating artificial general intelligence. (221 words)

## Articles & Blog Posts

- [docs/legal/end-user-license-agreement/index.html](/content/docs/legal/end-user-license-agreement/index.html) (1,442 words)
- [Manage models on Amazon EKS - Poolside](/content/docs/deployment/cloud/aws-eks/manage-models/index.html): Add, update, or remove inference models in an existing inference deployment on Amazon EKS. (673 words)
- [Relocate an on-premises server - Poolside](/content/docs/deployment/on-prem/relocation/index.html): Relocate a Poolside on-premises model inference server to a different physical network. (810 words)
- [Reference architecture - Poolside](/content/docs/deployment/cloud/aws-eks/reference-architecture/index.html): AWS reference architecture for a Poolside model inference deployment on Amazon EKS, including the architecture diagram, the AWS layers the inference chart depends on, and the key design decisions. (1,018 words)
- [Capacity planning - Poolside](/content/docs/deployment/capacity-planning/index.html): Use the capacity planner to estimate concurrent-agent capacity and developer seat counts for a Poolside deployment. (824 words)
- [Certified on-premises model inference stack: July 2026 - Poolside](/content/docs/deployment/on-prem/certified-stacks/july-2026/index.html): Certified software components and versions for the Poolside July 2026 on-premises model inference stack. (319 words)
- [Manage models on Kubernetes - Poolside](/content/docs/deployment/cloud/upstream-kubernetes/manage-models/index.html): Add, update, or remove inference models in an existing inference deployment on upstream Kubernetes. (693 words)
- [Use Poolside models in Pi - Poolside](/content/docs/tools/pi/index.html): Configure Pi to use Poolside models through Poolside Platform, a Poolside deployment, or OpenRouter., This guide shows how to set up pi.dev using Poolside through Poolside inference and then via Open Router. It begins by exporting a Poolside API key, using the latest configuration from docs.poolside.ai, and selecting the Laguna XS model to test a security vulnerability scan and a core config fix. Next, it switches to Open Router using the /slash login flow with an Open Router API key, selects the Open Router Laguna M model, and prompts pi to design and activate a custom Pi theme based on poolside.ai branding. The session ends confirming the theme is active with a quick greeting. (396 words)
- [Use Poolside models in OpenCode - Poolside](/content/docs/tools/opencode/index.html): Configure OpenCode to use Poolside models through Poolside Platform, a Poolside deployment, or OpenRouter., This guide explains how to set up Poolside models with OpenCode using Poolside Inference and then OpenRouter. First, it shows connecting in OpenCode via a slash connect command, selecting Poolside, generating a Poolside API key at platform.poolside.ai, and switching to the Laguna XS model to test codebase architecture and legacy C code vulnerabilities. It then uses the Laguna M model to inspect C code, fix a high severity buffer overflow, and validate that tests compile with all green. Next, it connects OpenCode to OpenRouter via slash connect, provides an OpenRouter API key, selects Poolside Laguna models, checks dependencies for outdated packages, updates Axum, and re-verifies compilation. (376 words)
- [Server and service maintenance - Poolside](/content/docs/deployment/on-prem/server-maintenance/index.html): Stop, start, reboot, or fully shut down a Poolside on-premises model inference node. (608 words)
- [Upgrade on Kubernetes - Poolside](/content/docs/deployment/cloud/upstream-kubernetes/upgrade/index.html): Upgrade an existing model inference deployment on upstream Kubernetes to a new bundle. (650 words)
- [Use Poolside models in Kilo Code - Poolside](/content/docs/tools/kilo-code/index.html): Configure Kilo Code to use Poolside models through Kilo Gateway, the Poolside Platform OpenAI-compatible endpoint, or OpenRouter., This guide shows how to set up Poolside models in Kilo Code using a Kilo Gateway, Open Router, or Poolside directly. It starts by installing Kilo Code from the Visual Studio Code Marketplace, verifying it responds, and using the free Poolside Laguna M model to explore an opened codebase and then run the app. It then covers configuring Kilo Code with an Open Router API key, selecting the free Laguna XS model to confirm the setup works. Finally, it explains setting up a custom Poolside provider via an OpenAI compatible inference endpoint from platform.poolside.ai using a generated Poolside API key and selecting available models. (404 words)
- [Use Poolside models in Cline - Poolside](/content/docs/tools/cline/index.html): Configure Cline to use Poolside models through Poolside Platform or OpenRouter., This Loom explains how to set up Cline with Poolside models using Poolside Inference and then using OpenRouter. It walks through configuring OpenAI compatible settings in Visual Studio Code by using the Poolside inference endpoint as the base URL and adding a Poolside API key, then selecting the Laguna XS model and updating model configuration. It demonstrates testing by starting a new chat, exploring a codebase, and switching to Laguna M to add a default fallback when a weather API is offline, validating that code compiles and tests stay green. Finally, it shows setting Cline to use OpenRouter, selecting a Poolside Laguna model, testing language terminology queries, and adding inline documentation after switching to Laguna M. (242 words)
- [Use Poolside models in Goose - Poolside](/content/docs/tools/goose/index.html): Configure Goose to use Poolside models through Poolside Platform, a Poolside deployment, or OpenRouter., This guide demonstrates how to set up Goose with Poolside using both OpenRouter and Poolside inference. It first installs Goose, selects the recommended OpenRouter option in the setup wizard, and approves an authorization request so no API key copying is needed, then uses /model commands to run system health checks, inspect network devices, and create a recurring daily check via a cron job. Next it covers Poolside inference by choosing Custom Provider with OpenAI Compatible, pasting the API endpoint from platform.poolside.ai, generating an API key for Goose, adding model IDs, enabling streaming, and using the V1 chat completions path for testing. It finishes with an example where Goose creates an automated daily news digest at 9 AM using cron and scripts to avoid repeats. (328 words)
- [Use Poolside models in GitHub Copilot - Poolside](/content/docs/tools/copilot/index.html): Configure GitHub Copilot in VS Code to use Poolside models through OpenRouter., This guide explains how to connect the Poolside Laguna model to Copilot in VS Code using Open Router. It instructs viewers to open Copilot chat, go to model management settings, select Add Models, and choose Open Router. The video then has you generate a dedicated API key on OpenRouter.ai named Poolside Copilot, copy it, and paste it into the VS Code setup bar. Finally, it directs viewers to select the Laguna XS2 or M1 model in the setup and run a quick test message to verify the connection, noting both models are free for a limited time. (203 words)
- [Use Poolside models in Hermes - Poolside](/content/docs/tools/hermes/index.html): Configure Hermes to use Poolside models through OpenRouter., This guide explains how to set up Poolside with Hermes through OpenRouter to manage a recurring desktop organization task. It covers running the Hermes install script, choosing full setup, selecting OpenRouter as the inference provider, and entering an OpenRouter API key from the API keys page. After selecting a Poolside model and finishing setup, it starts Hermes using Hermes-dash-t-u-i and sends a test message to verify the connection. The demo then prompts Hermes to organize desktop files, and sets it to run weekly using the macOS system notification system. (93 words)
- [docs/legal/data-processing-agreement/index.html](/content/docs/legal/data-processing-agreement/index.html) (5,339 words)
- [Use Poolside in Neovim with CodeCompanion - Poolside](/content/docs/tools/neovim/index.html): Configure Neovim with CodeCompanion to use Poolside through Agent Client Protocol. (117 words)
- [docs/support/support-portal/index.html](/content/docs/support/support-portal/index.html) (38 words)
- [Install on-premises - Poolside](/content/docs/deployment/on-prem/install/index.html): Install Poolside model inference on a single on-premises GPU host by using the Terraform installation bundle. (2,631 words)
- [Install on OpenShift - Poolside](/content/docs/deployment/cloud/openshift/install/index.html): Deploy Poolside model inference on OpenShift and serve models through an OpenAI-compatible API. (1,943 words)
- [System checks - Poolside](/content/docs/deployment/system-checks/index.html): Use dive to validate infrastructure readiness before and after deploying the Poolside inference stack, and to check GPU node configuration. (2,072 words)
- [Install on Kubernetes - Poolside](/content/docs/deployment/cloud/upstream-kubernetes/install/index.html): Deploy Poolside model inference on a self-managed Kubernetes cluster and serve models through an OpenAI-compatible API. (1,340 words)
- [Use LiteLLM with Poolside model inference - Poolside](/content/docs/deployment/inference-gateways/litellm/index.html): Deploy LiteLLM as an OpenAI-compatible gateway in front of Poolside model inference., This guide demonstrates two ways to deploy and test Light LLM with Ripple sign models. First, it deploys Light LLM locally with Docker Compose, then uses the Light LLM UI at /ui (port 4000) and an API endpoint on an OpenShift cluster to add a custom model named with the openai/ prefix (for example laguna s) and verify connectivity via UI and the harness. Next, it deploys Light LLM inside the Kubernetes cluster using Helm, creates required secrets for the master key and Light LLM Salt key, configures an OpenShift route without TLS, and confirms the migrations job completes and the endpoint responds. It then repeats model creation and testing using the in-cluster service URL and validates inference through both the UI and harness logs. (1,342 words)
- [Run Poolside in GitHub Actions - Poolside](/content/docs/tools/github-actions/index.html): Automate code review, security scans, release notes, and other repository tasks with Poolside in GitHub Actions workflows., This guide Loom demonstrates setting up Poolside automated code reviews and scheduled security scans with GitHub Actions. It shows generating a Poolside API key on platform.poolside.ai, adding it to GitHub repository secrets, and creating an Actions workflow that triggers on new pull requests. The workflow installs the Poolside agent CLI, runs a pool exec review against the origin branch using the Laguna model, and posts the resulting markdown review as a PR comment, then commits and tests the flow with a simulated PR containing backend changes. It then updates the workflow to run nightly via a Cron schedule and manually for the demo, scans the entire repository for vulnerabilities, outputs results to a JSON artifact, and explains how to download and triage the findings. (1,183 words)
- [Upgrade on Amazon EKS - Poolside](/content/docs/deployment/cloud/aws-eks/upgrade/index.html): Upgrade an existing model inference deployment on Amazon EKS to a new bundle. (671 words)
- [Install on Amazon EKS - Poolside](/content/docs/deployment/cloud/aws-eks/install/index.html): Deploy Poolside model inference on Amazon EKS with Helm, using IRSA for S3 access and an Application Load Balancer for ingress. (684 words)
- [Upgrade on-premises - Poolside](/content/docs/deployment/on-prem/upgrade/index.html): Upgrade an on-premises Poolside model inference deployment to a new installation bundle. (1,111 words)
- [Upgrade on OpenShift - Poolside](/content/docs/deployment/cloud/openshift/upgrade/index.html): Upgrade an existing model inference deployment on OpenShift to a new bundle. (692 words)
- [Documenting code - Poolside](/content/docs/resources/use-cases/documenting-code/index.html): This use case demonstrates how to use Poolside to generate, improve, and maintain code documentation. This can be especially useful when interacting with legacy codebases and understanding how existing codebases work. (1,190 words)
- [Manage models on OpenShift - Poolside](/content/docs/deployment/cloud/openshift/manage-models/index.html): Add, update, or remove inference models in an existing inference deployment on OpenShift. (695 words)
- [Inference metrics - Poolside](/content/docs/metrics/inference-metrics/index.html): Reference for the Prometheus metrics the Poolside vLLM model server exposes, and how to scrape them. (729 words)
- [Model release notes - Poolside](/content/docs/release-notes/models/index.html): Release notes for Poolside model releases. (659 words)
- [Refactoring code - Poolside](/content/docs/resources/use-cases/refactoring-code/index.html): This use case demonstrates how to use Poolside to refactor code by understanding existing implementations and suggesting improvements while maintaining functionality. (825 words)
- [Admin toolkit - Poolside](/content/docs/deployment/on-prem/admin/index.html): Common commands and checks for operating a Poolside on-premises model inference deployment. (732 words)
- [Remove from upstream Kubernetes - Poolside](/content/docs/deployment/cloud/upstream-kubernetes/remove/index.html): Instructions for removing a model inference deployment from an upstream Kubernetes cluster. (346 words)
- [Remove from Amazon EKS - Poolside](/content/docs/deployment/cloud/aws-eks/remove/index.html): Instructions for removing a model inference deployment from an Amazon EKS cluster. (426 words)
- [On-premises deployment FAQ - Poolside](/content/docs/deployment/on-prem/faq/index.html): Answers to common questions about Poolside on-premises model inference deployments. (749 words)
- [Remove from OpenShift - Poolside](/content/docs/deployment/cloud/openshift/remove/index.html): Instructions for removing a model inference deployment from Red Hat OpenShift. (345 words)
- [Videos - Poolside](/content/docs/resources/videos/index.html): Learn how to use Poolside with demos and walkthroughs., This guide demonstrates how to install and run our Laguna XS model locally on a Nvidia DGX-Bark GB10 without any cloud connectivity. Everything runs entirely on the box, with nothing leaving the machine., This Loom explains how to set up Cline with Poolside models using Poolside Inference and then using OpenRouter. It walks through configuring OpenAI compatible settings in Visual Studio Code by using the Poolside inference endpoint as the base URL and adding a Poolside API key, then selecting the Laguna XS model and updating model configuration. It demonstrates testing by starting a new chat, exploring a codebase, and switching to Laguna M to add a default fallback when a weather API is offline, validating that code compiles and tests stay green. Finally, it shows setting Cline to use OpenRouter, selecting a Poolside Laguna model, testing language terminology queries, and adding inline documentation after switching to Laguna M., This guide Loom demonstrates setting up Poolside automated code reviews and scheduled security scans with GitHub Actions. It shows generating a Poolside API key on platform.poolside.ai, adding it to GitHub repository secrets, and creating an Actions workflow that triggers on new pull requests. The workflow installs the Poolside agent CLI, runs a pool exec review against the origin branch using the Laguna model, and posts the resulting markdown review as a PR comment, then commits and tests the flow with a simulated PR containing backend changes. It then updates the workflow to run nightly via a Cron schedule and manually for the demo, scans the entire repository for vulnerabilities, outputs results to a JSON artifact, and explains how to download and triage the findings., This guide explains how to connect the Poolside Laguna model to Copilot in VS Code using Open Router. It instructs viewers to open Copilot chat, go to model management settings, select Add Models, and choose Open Router. The video then has you generate a dedicated API key on OpenRouter.ai named Poolside Copilot, copy it, and paste it into the VS Code setup bar. Finally, it directs viewers to select the Laguna XS2 or M1 model in the setup and run a quick test message to verify the connection, noting both models are free for a limited time., This guide demonstrates how to set up Goose with Poolside using both OpenRouter and Poolside inference. It first installs Goose, selects the recommended OpenRouter option in the setup wizard, and approves an authorization request so no API key copying is needed, then uses /model commands to run system health checks, inspect network devices, and create a recurring daily check via a cron job. Next it covers Poolside inference by choosing Custom Provider with OpenAI Compatible, pasting the API endpoint from platform.poolside.ai, generating an API key for Goose, adding model IDs, enabling streaming, and using the V1 chat completions path for testing. It finishes with an example where Goose creates an automated daily news digest at 9 AM using cron and scripts to avoid repeats., This guide explains how to set up Poolside with Hermes through OpenRouter to manage a recurring desktop organization task. It covers running the Hermes install script, choosing full setup, selecting OpenRouter as the inference provider, and entering an OpenRouter API key from the API keys page. After selecting a Poolside model and finishing setup, it starts Hermes using Hermes-dash-t-u-i and sends a test message to verify the connection. The demo then prompts Hermes to organize desktop files, and sets it to run weekly using the macOS system notification system., This guide shows how to set up Poolside models in Kilo Code using a Kilo Gateway, Open Router, or Poolside directly. It starts by installing Kilo Code from the Visual Studio Code Marketplace, verifying it responds, and using the free Poolside Laguna M model to explore an opened codebase and then run the app. It then covers configuring Kilo Code with an Open Router API key, selecting the free Laguna XS model to confirm the setup works. Finally, it explains setting up a custom Poolside provider via an OpenAI compatible inference endpoint from platform.poolside.ai using a generated Poolside API key and selecting available models., This guide explains how to set up Poolside models with OpenCode using Poolside Inference and then OpenRouter. First, it shows connecting in OpenCode via a slash connect command, selecting Poolside, generating a Poolside API key at platform.poolside.ai, and switching to the Laguna XS model to test codebase architecture and legacy C code vulnerabilities. It then uses the Laguna M model to inspect C code, fix a high severity buffer overflow, and validate that tests compile with all green. Next, it connects OpenCode to OpenRouter via slash connect, provides an OpenRouter API key, selects Poolside Laguna models, checks dependencies for outdated packages, updates Axum, and re-verifies compilation., This guide shows how to set up pi.dev using Poolside through Poolside inference and then via Open Router. It begins by exporting a Poolside API key, using the latest configuration from docs.poolside.ai, and selecting the Laguna XS model to test a security vulnerability scan and a core config fix. Next, it switches to Open Router using the /slash login flow with an Open Router API key, selects the Open Router Laguna M model, and prompts pi to design and activate a custom Pi theme based on poolside.ai branding. The session ends confirming the theme is active with a quick greeting., This guide demonstrates two ways to deploy and test Light LLM with Ripple sign models. First, it deploys Light LLM locally with Docker Compose, then uses the Light LLM UI at /ui (port 4000) and an API endpoint on an OpenShift cluster to add a custom model named with the openai/ prefix (for example laguna s) and verify connectivity via UI and the harness. Next, it deploys Light LLM inside the Kubernetes cluster using Helm, creates required secrets for the master key and Light LLM Salt key, configures an OpenShift route without TLS, and confirms the migrations job completes and the endpoint responds. It then repeats model creation and testing using the in-cluster service URL and validates inference through both the UI and harness logs. (151 words)
- [Amazon EKS deployment - Poolside](/content/docs/deployment/cloud/aws-eks/overview/index.html): Overview of deploying Poolside model inference on Amazon EKS by using Helm, with IRSA for object storage and an ALB for ingress. (413 words)
- [Certified stacks - Poolside](/content/docs/deployment/on-prem/certified-stacks/overview/index.html): Certified software stacks for supported Poolside on-premises model inference deployments. (158 words)
- [Submit a Support Request - Poolside](/content/docs/support/submit-a-support-request/index.html): How to open a support request through the Poolside support portal, set the right metadata, and write requests that get resolved faster. (386 words)
- [Visualizing data flows - Poolside](/content/docs/resources/use-cases/visualizing-data-flows/index.html): This use case demonstrates how to use Poolside to generate visual representations of code structures, workflows, and architectures using Mermaid diagrams. (345 words)
- [Prompting best practices - Poolside](/content/docs/resources/prompting-best-practices/index.html): Learn how to write effective prompts to get the most out of Poolside. (703 words)
- [Premium Support - Poolside](/content/docs/support/premium-support/index.html): Learn about Poolside Premium Support, response times, and scope. (162 words)
- [Trust Center - Poolside](/content/docs/legal/trust-center/index.html): Security, compliance, and privacy information is available in the Poolside Trust Center. (35 words)
- [Standard Support - Poolside](/content/docs/support/standard-support/index.html): Learn about Poolside Standard Support, response times, and more. (115 words)
- [Upstream Kubernetes deployment - Poolside](/content/docs/deployment/cloud/upstream-kubernetes/overview/index.html): Overview of deploying Poolside model inference on a self-managed Kubernetes cluster by using Helm. (169 words)
- [Tool integrations - Poolside](/content/docs/tools/index.html): Use Poolside with editors, desktop apps, coding agents, and local model runtimes. (177 words)
- [Use Poolside in another ACP-compatible editor - Poolside](/content/docs/tools/other-acp/index.html): Configure any ACP-compatible editor to use Poolside through Agent Client Protocol. (258 words)
- [How agents work - Poolside](/content/docs/resources/how-agents-work/index.html): A conceptual overview of agents powered by large language models, including their core components, behavior, and common use cases. (448 words)
- [Poolside Inference release notes - Poolside](/content/docs/release-notes/inference/index.html): Release notes for Poolside Inference deployment and runtime updates. (61 words)
- [Cloud deployment - Poolside](/content/docs/deployment/cloud/overview/index.html): Deploy Poolside model inference in a supported GPU-backed cloud Kubernetes environment. (219 words)
- [Use Poolside in Zed - Poolside](/content/docs/tools/zed/index.html): Configure Zed to use Poolside through Agent Client Protocol. (92 words)
- [Use Poolside in JetBrains - Poolside](/content/docs/tools/jetbrains/index.html): Configure JetBrains IDEs to use Poolside through Agent Client Protocol. (90 words)
- [Example workflows - Poolside](/content/docs/resources/example-workflows/index.html): Common tasks you can accomplish with Poolside Agent CLI. (73 words)
- [Use Poolside where you work - Poolside](/content/docs/use-poolside/index.html): Use Poolside models in your development workflow. (370 words)
- [trajectories/trials/019f2a95-b4b3-77b8-ad7c-dbdf59c0d9ca/download/index.html](/content/trajectories/trials/019f2a95-b4b3-77b8-ad7c-dbdf59c0d9ca/download/index.html) (64,690 words)
- [enterprise/index.html](/content/enterprise/index.html) (261 words)
- [Poolside Agent CLI - Poolside](/content/docs/cli/pool/index.html): Use the pool command for interactive sessions, automated tasks, ACP editors, and other agent servers. (411 words)
- [OpenAI-compatible API examples - Poolside](/content/docs/api/openai-api-examples/index.html): Make OpenAI-compatible API requests to Poolside models. (937 words)
- [docs/llms-txt.html](/content/docs/llms-txt.html) (1,265 words)
- [CLI reference - Poolside](/content/docs/cli/cli-reference/index.html): Reference for current pool commands, flags, slash commands, and exit codes. (2,224 words)
- [Work from the terminal - Poolside](/content/docs/cli/interactive-mode/index.html): Use pool to work with a Poolside agent over multiple turns in the terminal. (2,345 words)
- [Settings file reference - Poolside](/content/docs/settings-file-reference/index.html): Reference for settings you can define in Poolside settings.yaml files. (1,049 words)
- [Set up offline documentation - Poolside](/content/docs/deployment/cloud/set-up-offline-documentation/index.html): Deploy the Poolside documentation site in-cluster for offline access as part of a cloud inference deployment. (714 words)
- [STIG hardening considerations - Poolside](/content/docs/deployment/on-prem/stig/index.html): Considerations for deploying Poolside on-premises model inference on STIG-hardened systems, including known compatibility issues and recommended configurations. (599 words)
- [Poolside API - Poolside](/content/docs/api/overview/index.html): Get started with the Poolside OpenAI-compatible API: base URLs, keys, and your first request. (746 words)
- [On-premises deployment - Poolside](/content/docs/deployment/on-prem/overview/index.html): Overview of Poolside on-premises model inference deployments, including hardware options, architecture, installation approach, and operational considerations. (713 words)
- [Permissions - Poolside](/content/docs/permissions/index.html): Control what Poolside agents can read, write, and run by using settings.yaml. (1,306 words)
- [Storage requirements - Poolside](/content/docs/deployment/on-prem/storage/index.html): Storage guidance for Poolside on-premises model deployments, including partitioning recommendations and model checkpoint storage. (425 words)
- [Skills - Poolside](/content/docs/skills/index.html): Use local skills to give agents reusable, task-specific instructions and resources. (908 words)
- [Install Poolside Agent CLI - Poolside](/content/docs/cli/install/index.html): Install pool and sign in. (1,199 words)
- [MCP servers - Poolside](/content/docs/mcp-servers/index.html): Configure personal MCP servers that Poolside agents can invoke from the CLI. (1,025 words)
- [OpenShift deployment - Poolside](/content/docs/deployment/cloud/openshift/overview/index.html): Overview of deploying Poolside model inference on Red Hat OpenShift by using Helm. (185 words)
- [Poolside Documentation - Poolside](/content/docs/index/index.html): Build with Poolside agentic coding models in your terminal, editor, tools, or applications. (64 words)
- [Trademark Guidelines — Poolside](/content/legal/trademark-guidelines/index.html): Legal information and policies for Poolside, Legal information and policies for Poolside (1,332 words)
- [Example prompts - Poolside](/content/docs/resources/example-prompts/index.html): Common development tasks and example prompts you can use with Poolside. (306 words)
- [AGENTS.md instructions - Poolside](/content/docs/agent-instructions/index.html): Set up instructions for Poolside across personal, project, and directory scopes. (815 words)
- [Troubleshoot Poolside Agent CLI - Poolside](/content/docs/cli/troubleshooting/index.html): Diagnose common Poolside Agent CLI authentication, configuration, and ACP issues. (697 words)
- [Sandboxes - Poolside](/content/docs/sandboxes/index.html): Run Poolside agents in local sandbox environments with file system and network controls. (792 words)
- [Web search and fetch - Poolside](/content/docs/web-search-and-fetch/index.html): Let Poolside agents search the web and read pages with the built-in web_search and web_fetch tools. (439 words)
- [Support Overview - Poolside](/content/docs/support/overview/index.html): Overview of Poolside support offerings, eligibility requirements, and options for non-standard deployments (629 words)
- [Automate tasks - Poolside](/content/docs/cli/automated-mode/index.html): Use pool exec to run one-shot tasks in scripts, CI, and automated workflows. (444 words)
- [Quickstart - Poolside](/content/docs/get-started/quickstart/index.html): Install Poolside Agent CLI, log in, and run your first agent task. (373 words)
- [Supported models - Poolside](/content/docs/get-started/supported-models/index.html): Compare supported Poolside models by intended use, context window, and reasoning support. (433 words)
- [Use other agent servers - Poolside](/content/docs/cli/other-agent-servers/index.html): Use pool as an Agent Client Protocol (ACP) client for another compatible agent server. (154 words)
- [Release notes - Poolside](/content/docs/release-notes/overview/index.html): Recent Poolside inference deployment and model releases. (283 words)
- [Poolside overview - Poolside](/content/docs/get-started/overview/index.html): Poolside provides agentic coding models, self-managed inference, APIs, and agent workflows for secure software engineering. (392 words)
- [Supported configurations - Poolside](/content/docs/deployment/supported-configurations/index.html): Review supported environments and minimum requirements for Poolside model inference. (271 words)
- [Configure Poolside - Poolside](/content/docs/configure/index.html): Shape agent behavior and control what Poolside agents can access. (182 words)
- [How to run a Poolside model locally - Poolside](/content/docs/resources/run-model-locally/index.html): Run an open-weight Poolside model locally to power a coding agent on your own hardware., This guide demonstrates how to install and run our Laguna XS model locally on a Nvidia DGX-Bark GB10 without any cloud connectivity. Everything runs entirely on the box, with nothing leaving the machine. (1,125 words)
- [Resources - Poolside](/content/docs/resources/overview/index.html): Find conceptual guides, task-focused how-tos, example prompts, and use cases for working with Poolside. (133 words)
- [Trajectories — Poolside](/content/trajectories/trials/019f7297-e7f9-7756-9644-6aadb6e2b668/index.html): Explore trajectories from the evaluation of Poolside models as they are released. (24,044 words)
- [Trajectories — Poolside](/content/trajectories/trials/019f2a95-b4b3-77b8-ad7c-dbdf59c0d9ca/index.html): Explore trajectories from the evaluation of Poolside models as they are released. (7,226 words)
- [Tools of the Trade: C2C Activation Offloading on Grace Blackwell — Poolside](/content/blog/tools-of-the-trade-c2c-activation-offloading-on-grace-blackwell.html): We demonstrate the potential of NVIDIA's NVLink C2C on Grace-based superchips as a high-performance alternative to selective activation checkpointing. By offloading MLP activations to host memory during training, we achieve a 6–13% throughput improvement over selective AC with negligible memory overhead., We demonstrate the potential of NVIDIA's NVLink C2C on Grace-based superchips as a high-performance alternative to selective activation checkpointing. By offloading MLP activations to host memory during training, we achieve a 6–13% throughput improvement over selective AC with negligible memory overhead. (4,730 words)
- [Data Processing Addendum — Poolside](/content/legal/dpa/index.html): Legal information and policies for Poolside, Legal information and policies for Poolside (5,356 words)
- [Poolside Terms of Use — Poolside](/content/legal/eula/index.html): Legal information and policies for Poolside, Legal information and policies for Poolside (3,796 words)
- [The carrier and the beacon — Poolside](/content/blog/the-carrier-and-the-beacon/index.html): Running inference and evaluations inside the Model Factory, Running inference and evaluations inside the Model Factory (5,294 words)
- [The hidden engineering behind foundation model building — Poolside](/content/blog/introducing-the-model-factory/index.html): Introducing the Model Factory, Introducing the Model Factory (4,603 words)
- [Designing a world-class code execution environment — Poolside](/content/blog/designing-a-world-class-code-execution-environment/index.html): We present the details of our code execution platform, We present the details of our code execution platform (4,136 words)
- [Titan, the Model Factory's furnace — Poolside](/content/blog/titan-the-model-factory-s-furnace/index.html): We present details of our distributed training codebase, known as Titan., We present details of our distributed training codebase, known as Titan. (4,132 words)
- [Poolside Privacy Policy — Poolside](/content/legal/privacy/index.html): Legal information and policies for Poolside, Legal information and policies for Poolside (3,188 words)
- [Secrets - Poolside](/content/docs/secrets/index.html): Store sensitive values like API keys and tokens that agents can use without exposing the raw value. (568 words)
- [Member of Engineering (Design Engineer, Product) — Poolside](/content/careers/member-of-engineering-design-engineer-product-7fb2ceb5-6ace-497f-9c8b-124a9317f789.html): Join us at Poolside and work at the forefront of applied research and engineering at scale., Join us at Poolside and work at the forefront of applied research and engineering at scale. (1,494 words)
- [Member of Security (Corporate Technology) — Poolside](/content/careers/member-of-security-corporate-technology-97536454-0ca9-40b9-8b68-3d63a040286c.html): Join us at Poolside and work at the forefront of applied research and engineering at scale., Join us at Poolside and work at the forefront of applied research and engineering at scale. (811 words)
- [Member of Engineering (Pre-training / Data Engineering) — Poolside](/content/careers/member-of-engineering-pre-training-data-engineering-ac8683a2-4b78-4d10-a528-fa6524548d90.html): Join us at Poolside and work at the forefront of applied research and engineering at scale., Join us at Poolside and work at the forefront of applied research and engineering at scale. (1,388 words)
- [Through the looking glass of benchmark hacking — Poolside](/content/blog/through-the-looking-glass/index.html): Outlining some of the reward hacks we’ve encountered and what strategies we are exploring to resolve them. , Outlining some of the reward hacks we’ve encountered and what strategies we are exploring to resolve them.  (1,858 words)
- [Research — Poolside](/content/research/index.html): Traditional foundation model training is manual, linear, and slow. We built something different., Traditional foundation model training is manual, linear, and slow. We built something different. (332 words)
- [The finishing touches — Poolside](/content/blog/post-training-in-the-model-factory/index.html): Using the Model Factory and its components to conduct post-training at scale, Using the Model Factory and its components to conduct post-training at scale (2,014 words)
- [Member of Engineering (Pre-training / Data Research) — Poolside](/content/careers/member-of-engineering-pre-training-data-research-819757ab-8e03-4c7b-8fed-be7451372cb4.html): Join us at Poolside and work at the forefront of applied research and engineering at scale., Join us at Poolside and work at the forefront of applied research and engineering at scale. (826 words)
- [Introducing Laguna XS.2 and Laguna M.1 — Poolside](/content/blog/introducing-laguna-xs2-m1/index.html): We’re releasing two foundation models and two products into preview today. , We’re releasing two foundation models and two products into preview today.  (1,062 words)
- [Member of Engineering (Synthetic Data Research) — Poolside](/content/careers/member-of-engineering-synthetic-data-research-f99aaa7d-473b-4710-8d2d-5f9893fb5b15.html): Join us at Poolside and work at the forefront of applied research and engineering at scale., Join us at Poolside and work at the forefront of applied research and engineering at scale. (909 words)
- [Member of Engineering (Infrastructure) — Poolside](/content/careers/member-of-engineering-infrastructure-ade02c95-890f-4f1d-9ca6-05076b6fe687.html): Join us at Poolside and work at the forefront of applied research and engineering at scale., Join us at Poolside and work at the forefront of applied research and engineering at scale. (745 words)
- [Member of Engineering (Evaluations / Engineering) — Poolside](/content/careers/member-of-engineering-evaluations-engineering-df54a608-9fd9-4c09-aad1-7de71a0d487e.html): Join us at Poolside and work at the forefront of applied research and engineering at scale., Join us at Poolside and work at the forefront of applied research and engineering at scale. (528 words)
- [Member of Engineering (Design Engineer, Full Experience) — Poolside](/content/careers/member-of-engineering-design-engineer-full-experience-0e5ec752-201c-4932-8750-5af1607e3ec4.html): Join us at Poolside and work at the forefront of applied research and engineering at scale., Join us at Poolside and work at the forefront of applied research and engineering at scale. (860 words)
- [Trajectories — Poolside](/content/trajectories/trials/019f6c37-7621-7ec1-96b5-f041f83c1540/index.html): Explore trajectories from the evaluation of Poolside models as they are released. (35,600 words)
- [Events Lead — Poolside](/content/careers/events-lead-9483625e-fa67-4c29-9150-e374f6cb4cdb/index.html): Join us at Poolside and work at the forefront of applied research and engineering at scale., Join us at Poolside and work at the forefront of applied research and engineering at scale. (1,075 words)
- [Member of Engineering (Post-training) — Poolside](/content/careers/member-of-engineering-post-training-7662c253-27bf-4a64-a52a-ecd221887993.html): Join us at Poolside and work at the forefront of applied research and engineering at scale., Join us at Poolside and work at the forefront of applied research and engineering at scale. (714 words)
- [Unveiling Poolside's first-party partnership with AWS — Poolside](/content/blog/unveiling-our-partnership-with-aws/index.html): Today, we’re thrilled to announce a transformative partnership with Amazon Web Services (AWS) that establishes Poolside as a first-party AWS offering., Today, we’re thrilled to announce a transformative partnership with Amazon Web Services (AWS) that establishes Poolside as a first-party AWS offering. (717 words)
- [Models — Poolside](/content/models/index.html): Poolside trains models from scratch—our own data, our own infrastructure, our own reinforcement learning. Competitive with leading models at a fraction of the compute. Dive into how we build foundation models from scratch., Poolside trains models from scratch—our own data, our own infrastructure, our own reinforcement learning. Competitive with leading models at a fraction of the compute. Dive into how we build foundation models from scratch. (746 words)
- [Member of Engineering (Experiment Platform) — Poolside](/content/careers/member-of-engineering-experiment-platform-fbedf9a3-d3f4-4311-8ffd-70def97a5b62.html): Join us at Poolside and work at the forefront of applied research and engineering at scale., Join us at Poolside and work at the forefront of applied research and engineering at scale. (684 words)
- [Our vision: Research — Poolside](/content/vision/research/index.html): When we founded Poolside in San Francisco in April 2023, the narrative in the industry was that all we needed to reach AGI was to scale up language modelling., When we founded Poolside in San Francisco in April 2023, the narrative in the industry was that all we needed to reach AGI was to scale up language modelling. (1,261 words)
- [Poolside at AWS re:Invent 2025](/content/events/aws-reinvent-2025/index.html): Poolside will showcase how sovereign, infrastructure-first AI can empower government, education, and public sector organizations. See how our foundation models run entirely within your environment—delivering intelligence without compromising security or control. , Poolside will showcase how sovereign, infrastructure-first AI can empower government, education, and public sector organizations. See how our foundation models run entirely within your environment—delivering intelligence without compromising security or control.  (363 words)
- [Member of Engineering (Agent Sandboxes) — Poolside](/content/careers/member-of-engineering-agent-sandboxes-9a4f25e5-d387-46b6-8c0f-ebc8b8837de6.html): Join us at Poolside and work at the forefront of applied research and engineering at scale., Join us at Poolside and work at the forefront of applied research and engineering at scale. (713 words)
- [Poolside on Dell: an efficient path for frontier AI inside your boundary — Poolside](/content/blog/poolside-on-dell/index.html): Poolside running on Dell gives enterprises another path: frontier AI that runs inside their own boundary, closer to their data, with stronger governance, better performance, and more predictable economics., Poolside running on Dell gives enterprises another path: frontier AI that runs inside their own boundary, closer to their data, with stronger governance, better performance, and more predictable economics. (1,526 words)
- [Introducing Laguna XS 2.1 — Poolside](/content/blog/introducing-laguna-xs-2-1/index.html): Today we're releasing Laguna XS 2.1, an upgraded version of our Laguna XS.2 model., Today we're releasing Laguna XS 2.1, an upgraded version of our Laguna XS.2 model. (806 words)
- [Announcing our $500 million fundraise to make progress towards AGI — Poolside](/content/blog/announcing-our-500-million-fundraise-to-make-progress-towards-agi.html): AI has the potential to drive abundance for humanity in the 21st century at a magnitude larger than the last 100 years of technological progress., AI has the potential to drive abundance for humanity in the 21st century at a magnitude larger than the last 100 years of technological progress. (1,001 words)
- [Member of Engineering (Multimodality - Research Lead) — Poolside](/content/careers/member-of-engineering-multimodality-research-lead-0d16412c-58a2-464c-8407-6413210a8c72.html): Join us at Poolside and work at the forefront of applied research and engineering at scale., Join us at Poolside and work at the forefront of applied research and engineering at scale. (987 words)
- [Citigroup’s Global Technology Banking Chief Philip Drury Joins AI company Poolside as Chief Investment Officer — Poolside](/content/newsroom/philip-drury-joins-poolside-as-chief-investment-officer.html): Poolside is a foundation model company bringing intelligence to everywhere work gets done. Our mission is to drive abundance for humanity by creating artificial general intelligence. (417 words)
- [Blog — Poolside](/content/blog/index.html): Thoughts from the team building the foundation models and platform redefining how software gets made—securely, intelligently, and at scale., Thoughts from the team building the foundation models and platform redefining how software gets made—securely, intelligently, and at scale. (584 words)
- [Member of Engineering (Evaluations) — Poolside](/content/careers/member-of-engineering-evaluations-ba11fe78-f6f6-4165-b76b-020a46ad8fee.html): Join us at Poolside and work at the forefront of applied research and engineering at scale., Join us at Poolside and work at the forefront of applied research and engineering at scale. (454 words)
- [Laguna XS.2 and M.1: A Deeper Dive — Poolside](/content/blog/laguna-a-deeper-dive/index.html): We've released the first two models in the Laguna family, Laguna M.1 and Laguna XS.2, alongside the runtime we use to train and operate agents, available through two product experiences in research preview. , We've released the first two models in the Laguna family, Laguna M.1 and Laguna XS.2, alongside the runtime we use to train and operate agents, available through two product experiences in research preview.  (847 words)
- [Newsroom — Poolside](/content/newsroom/index.html): News and narratives from the frontier of AI., News and narratives from the frontier of AI. (116 words)
- [Long context update: Laguna XS.2 and M.1 — Poolside](/content/blog/long-context-update-laguna-xs-2-and-m-1/index.html): Now serving our Laguna XS.2 and M.1 models at 256K context lengths., Now serving our Laguna XS.2 and M.1 models at 256K context lengths. (405 words)
- [Introducing Laguna S 2.1 — Poolside](/content/blog/introducing-laguna-s-2-1/index.html): Today we’re releasing Laguna S 2.1, a significant step forward in our development of models that pursue longer horizon work and make effective use of reasoning., Today we’re releasing Laguna S 2.1, a significant step forward in our development of models that pursue longer horizon work and make effective use of reasoning. (623 words)
- [Announcing the acquisition of Fern Labs — Poolside](/content/blog/fern-labs-acquisition/index.html): We’re excited to announce that Poolside acquired Fern Labs, a London-based forward‑deployed research engineering company behind Bridge, a multi‑agent orchestration layer built for high‑stakes production environments., We’re excited to announce that Poolside acquired Fern Labs, a London-based forward‑deployed research engineering company behind Bridge, a multi‑agent orchestration layer built for high‑stakes production environments. (600 words)
- [Our vision: Purpose — Poolside](/content/vision/purpose/index.html): We believe in a future where AI drives abundance—making science, health, housing, food, and education better, cheaper, and more accessible., We believe in a future where AI drives abundance—making science, health, housing, food, and education better, cheaper, and more accessible. (710 words)
- [AI, your way: introducing the Poolside Platform — Poolside](/content/blog/introducing-the-poolside-platform/index.html): Production-grade AI agents deployed inside your security boundary—with full auditability, governance, and no cloud compromise., Production-grade AI agents deployed inside your security boundary—with full auditability, governance, and no cloud compromise. (501 words)
- [Member of Engineering (Compute) — Poolside](/content/careers/member-of-engineering-compute-1f2a733a-a3fe-48fe-bf61-1425eddb30f6.html): Join us at Poolside and work at the forefront of applied research and engineering at scale., Join us at Poolside and work at the forefront of applied research and engineering at scale. (654 words)
- [Enabling the Agentic Enterprise with Redpanda — Poolside](/content/blog/partnering-with-redpanda/index.html): Our strategic partnership with Redpanda's Agentic Data Plane, Our strategic partnership with Redpanda's Agentic Data Plane (280 words)
- [Get Started — Poolside](/content/get-started/index.html): Poolside is a foundation model company bringing intelligence to everywhere work gets done. Our mission is to drive abundance for humanity by creating artificial general intelligence., Poolside is a foundation model company bringing intelligence to everywhere work gets done. Our mission is to drive abundance for humanity by creating artificial general intelligence. (171 words)
- [Careers — Poolside](/content/careers/index.html): Join our mission at Poolside and work alongside some of the world's top talent, at the forefront of applied research and engineering at scale., Join our mission at Poolside and work alongside some of the world's top talent, at the forefront of applied research and engineering at scale. (202 words)
- [Events — Poolside](/content/events/index.html): Meet our team and learn how to power your organization with foundational AI., Meet our team and learn how to power your organization with foundational AI. (66 words)
- [assets/laguna/laguna-m1-xs2-technical-report-pdf.html](/content/assets/laguna/laguna-m1-xs2-technical-report-pdf.html) (18,408 words)
- [Government — Poolside](/content/government/index.html): Deploy Poolside inside your own environment, maintain full data sovereignty, and operate with strict security and compliance controls., Deploy Poolside inside your own environment, maintain full data sovereignty, and operate with strict security and compliance controls. (846 words)

## About Pages

- [Talk to Poolside today to start your journey — Poolside](/content/contact/index.html): The future of software starts with owning your models., The future of software starts with owning your models. (38 words)

## Resources

- [Full Page Index](/index.html): Browse all cached pages with rich metadata
- [About This Cache](/content/about.html): Methodology, technical details, and usage guidelines
- [XML Sitemap](/sitemap.xml): Machine-readable sitemap for crawler discovery
- [Robots.txt](/robots.txt): Crawler directives