Valuation & Funding
On November 4, 2026, OpenAI announced that it will acquire Pierre Computer Company, the maker of Code Storage and the open-source Trees and Diffs libraries. The price was not disclosed. Closing is subject to customary conditions, including regulatory approvals, and until then the two remain separate, independent companies.
Product
Pierre sells Code Storage, a hosted Git service that software companies build into their own products. The typical buyer is a developer team at an AI coding company that needs a place to keep the code, branches and history its agents and users generate.
Code Storage
Git is the system programmers use to record a project's history. Each commit is a saved snapshot of the files, and each branch is a separate line of work that can later be merged. Pierre argues that teams building AI apps otherwise stitch together file storage, databases and GitHub sync jobs that are slow and brittle.
A developer adds Pierre's TypeScript, Python or Go SDK, authenticates with a private key and creates a repository with one API call. The app gets back a Git address, which can sit on the app's own domain, so its users run ordinary clone and push commands without seeing Pierre.
For example, an application could let an agent write commits, open short-lived "ephemeral branches" to test competing fixes in parallel, save its memory files and merge the winning attempt. The app can stream files, search code and receive webhooks that alert build systems or bots. Repositories can also sync with GitHub, GitLab and Bitbucket.
Pierre advertises no caps on storage or request rates, repositories up to 32TB, a 99.99% uptime SLA, and SOC 2 Type II and HIPAA compliance. Each write is confirmed across several replicas, so copies stay consistent through server failures. Repositories untouched for seven days move to compressed cold storage.
Open-source libraries
Diffs displays code and side-by-side comparisons of what changed between versions. It includes comments, accept/reject buttons, merge-conflict resolution and an edit mode for fixing a line directly inside the diff. Trees displays a project's folder structure with Git status badges and search, and it renders fast enough to handle tens of thousands of files.
Both libraries already power file browsing and change review in OpenAI's Codex app. Pierre also lists DiffsHub among its open-source projects and recently released Highlights, a syntax highlighter designed to update as developers type or agents stream code.
Business Model
Code Storage customers receive a monthly usage bill. Hot storage costs $0.005 per GB per hour with replication included, and cold storage costs $0.0002 per GB per hour. It also meters inbound pushes and outbound clones and fetches.
Plans set a minimum monthly spend rather than an extra fee. Starter begins at $20, and Scale begins at $200 with single-tenant options, SSO and 24/7 support. Enterprise adds private networking, volume discounts and custom commitments.
Pierre pays for the cloud storage, replicas and network traffic behind each repository. Pierre says cold storage is priced at cost, making the mix of active and archived repositories important to profitability. Actual margins are undisclosed, as is pricing for Managed Code Storage running in customers' own clouds.
Pierre distributes its libraries as open-source software, separately from the metered Code Storage service. Their adoption, including inside Codex, gives developers a way to encounter Pierre’s tools, but a library installation is not evidence of a paying storage customer.
Competition
Code Storage is infrastructure a company embeds inside its own app, while its best-known rivals sell complete collaboration platforms that developers log into directly. The contest is whether AI coding companies keep agent code on GitHub or GitLab, build their own Git backends as OpenAI did with its in-house Pando system, or rent a white-label layer like Code Storage.
GitHub
GitHub, owned by Microsoft, bundles Git hosting with pull requests, code review, Actions for automated builds, Codespaces, Copilot and Advanced Security scanning. It also offers APIs, webhooks and enterprise identity controls. Its advantage is incumbency: teams already collaborating on GitHub can automate through accounts they have.
Pierre syncs with GitHub rather than replacing it outright, and its pitch is control. With Code Storage, the app owner controls the repository experience and avoids the rate limits, complex authorization and Microsoft API-access risk that Pierre says buyers worry about.
GitLab
GitLab sells itself as one control plane for teams and agents, combining source control, merge requests, CI/CD, security scanning, compliance and its Duo Agent Platform. It offers multi-tenant SaaS, a self-managed version, single-tenant Dedicated and a FedRAMP Moderate government edition. GitLab Flex spreads one annual commitment across seats and usage credits.
GitLab integrates Codex, Claude Code and Cursor while aiming to remain the governed system of record. Code Storage competes lower in the stack. It has no planning, CI or security suite, but its API is built for creating repositories at machine scale inside someone else's product.
TAM Expansion
Pierre's core market is AI coding products that need Git storage behind their apps. Its documentation and enterprise plans point to two adjacent buyers, though neither is disclosed as an existing customer base. OpenAI's plan to bring Code Storage into Codex after closing could also reshape both paths.
Agent-development infrastructure
Pierre's docs describe saving agent memory, capturing intermediate agent state and isolating parallel attempts. Its site also says it stores other machine-generated artifacts beyond code. If general-purpose agent builders adopt Git as a versioned store for agent output, Code Storage could draw on AI infrastructure budgets rather than developer-tools budgets.
Pierre's edge would be Git's snapshots, which can be pinned, compared and rolled back, and which Pierre argues suit prompts and evaluations. Entering this market would mean selling to agent-framework and sandbox providers, with revenue coming from the same storage and traffic meters.
Enterprise-managed deployments
Managed Code Storage lets customers bring their own cloud, and the Enterprise plan adds single-tenant or private networking. A regulated company's platform team building internal coding agents could buy it to keep agent-generated code inside its own environment.
That would tap security and infrastructure budgets through longer committed contracts, but it would require enterprise-grade support and deployment tooling. It also depends on OpenAI continuing to sell a standalone, self-deployed product rather than steering these buyers into Codex.
Risks
Acquisition and neutrality. The deal awaits regulatory approval, and once it closes OpenAI plans to bring Code Storage's features into Codex. AI coding companies that compete with Codex may hesitate to keep their code on a rival's infrastructure.
Low-margin archive growth. Pierre prices cold storage at cost and moves repositories there after seven idle days. If agents create mostly short-lived repositories, billed storage could grow without matching gross profit.
Build-or-bundle substitutes. GitHub and GitLab bundle Git storage with collaboration and CI tools that buyers already pay for. OpenAI built Pando in-house before announcing this deal, which shows the largest agent platforms can build their own Git backends.
DISCLAIMERS
This report is for information purposes only and is not to be used or considered as an offer or the solicitation of an offer to sell or to buy or subscribe for securities or other financial instruments. Nothing in this report constitutes investment, legal, accounting or tax advice or a representation that any investment or strategy is suitable or appropriate to your individual circumstances or otherwise constitutes a personal trade recommendation to you.
This research report has been prepared solely by Sacra and should not be considered a product of any person or entity that makes such report available, if any.
Information and opinions presented in the sections of the report were obtained or derived from sources Sacra believes are reliable, but Sacra makes no representation as to their accuracy or completeness. Past performance should not be taken as an indication or guarantee of future performance, and no representation or warranty, express or implied, is made regarding future performance. Information, opinions and estimates contained in this report reflect a determination at its original date of publication by Sacra and are subject to change without notice.
Sacra accepts no liability for loss arising from the use of the material presented in this report, except that this exclusion of liability does not apply to the extent that liability arises under specific statutes or regulations applicable to Sacra. Sacra may have issued, and may in the future issue, other reports that are inconsistent with, and reach different conclusions from, the information presented in this report. Those reports reflect different assumptions, views and analytical methods of the analysts who prepared them and Sacra is under no obligation to ensure that such other reports are brought to the attention of any recipient of this report.
All rights reserved. All material presented in this report, unless specifically indicated otherwise is under copyright to Sacra. Sacra reserves any and all intellectual property rights in the report. All trademarks, service marks and logos used in this report are trademarks or service marks or registered trademarks or service marks of Sacra. Any modification, copying, displaying, distributing, transmitting, publishing, licensing, creating derivative works from, or selling any report is strictly prohibited. None of the material, nor its content, nor any copy of it, may be altered in any way, transmitted to, copied or distributed to any other party, without the prior express written permission of Sacra. Any unauthorized duplication, redistribution or disclosure of this report will result in prosecution.