GenAI PMGBrain
A product or system referenced as a retrieval and latency data task target. The newsletter uses it as an example of where strong evaluations around retrieval and cost matter.
Key Highlights
- GBrain is positioned as an MIT-licensed open-source retrieval and memory system for AI agents and company-brain workflows.
- Newsletter coverage ties GBrain closely to retrieval quality, latency, and cost-per-compute evaluation discipline.
- It was described as state-of-the-art for agentic retrieval without LLM query rewriting, with public eval references in gbrain-evals.
- GBrain integrates with agent ecosystems including OpenClaw, Hermes, MCP, Codex, and Claude Code.
- Its evolution from retrieval layer to modular company-brain framework makes it relevant for PMs designing enterprise knowledge assistants.
At a glance
- Category
- Knowledge Management
GBrain
Overview
GBrain is an MIT-licensed open-source retrieval and memory system for AI agents, repeatedly described in the newsletter as a retrieval layer, retrieval engine, and broader “company brain” framework. Across mentions, it is positioned as infrastructure for agentic knowledge access: a system built to retrieve the right information from large markdown-based knowledge bases, support scoped AI workflows, and integrate with agent stacks such as OpenClaw, Hermes Agent, Codex, and Claude Code.For AI Product Managers, GBrain matters because it sits at the intersection of several practical product concerns: retrieval quality, latency, cost per compute, agent memory, and evaluation rigor. It is used in the newsletter as both a concrete OSS tool and an example of why strong evals around retrieval and latency matter when building agentic products that depend on large internal knowledge stores or “company brain” systems.
Key Developments
- 2026-05-20: Garry Tan released GBrain as an MIT-licensed open-source retrieval and memory system, saying it achieved state-of-the-art results on LongMemEval and outperformed known open-source repos by over 1% without LLM query rewriting.
- 2026-05-24: GBrain was described as an MIT-licensed, state-of-the-art retrieval engine for agents, built for OpenClaw and Hermes with MCP server support for broader agent-harness integration.
- 2026-05-24: An update added synthesized answers in addition to basic retrieval, with an A/B framing of “GBrain Search” vs. “GBrain Think.”
- 2026-06-01: Garry Tan open-sourced GBrain on GitHub and outlined a 30-minute setup around a 350k-page markdown LLM wiki plus an OpenClaw/Hermes agent workflow.
- 2026-06-06: GBrain was presented as a modular “company brain” framework organizing work through scoped AI agents and client pods, extending its positioning beyond retrieval alone.
- 2026-06-16: Garry Tan described GBrain as an MIT-licensed retrieval layer for AI agents, likening it to a just-in-time “3-book librarian,” and said it was already used by thousands of developers.
- 2026-07-21: Garry Tan formally launched GBrain as a free open-source retrieval library optimized for Hermes Agent and OpenClaw, powering his personal company brain and AI.
- 2026-07-28: Garry Tan said GBrain had become state-of-the-art for agentic retrieval without LLM rewriting, with evaluation results published in the gbrain-evals GitHub repo.
- 2026-08-13: GBrain v0.45.6.0 was released with 17 new brain skills, hardened using a personal OpenClaw agent over hundreds of thousands of markdown files; support for Codex and Claude Code was highlighted.
- 2026-08-18: The gbrain GitHub repo was shared as a private GitHub repo containing 70 “proven skills” and the beginnings of a “Karpathy-style knowledge wiki,” with MIT-licensed open-source materials and README documentation.
- 2026-10-11: GBrain was referenced as an example workload for why clear evaluations around retrieval, latency, and cost per compute matter, alongside the release of the gbrain-evals repository with 40+ categories.
Relevance to AI PMs
- Use it as a retrieval benchmark reference: GBrain is repeatedly tied to retrieval performance claims, LongMemEval, and agentic retrieval without query rewriting. PMs can use it as a reference point when scoping internal search, memory, or RAG features and when asking vendors or engineers for proof of retrieval quality.
- Study its evaluation posture, not just its features: The newsletter connects GBrain directly to evaluation needs across retrieval, latency, and cost per compute. For PMs, the tactical lesson is to define measurable eval categories before scaling an internal knowledge assistant or agent memory product.
- Model architecture decisions for company-brain products: GBrain is described both as a retrieval layer and as a modular company-brain framework with scoped agents and client pods. PMs building enterprise AI assistants can use this as a pattern for separating knowledge retrieval, orchestration, and task-specific agent workflows.
Related
- garry-tan: Primary creator and most frequent source of GBrain announcements, benchmarks, and repository shares.
- openclaw and hermes-agent / hermes: Core agent environments GBrain is repeatedly described as being built for or optimized around.
- codex and claude-code: Developer-agent tools that GBrain later added support for, broadening its practical integration surface.
- gbrain-evals: Companion evaluation repository used to frame retrieval, latency, and cost measurement for GBrain-like tasks.
- rag: GBrain is highly relevant to retrieval-augmented systems, especially where long-context knowledge access and memory quality matter.
- mcp: Mentioned as part of GBrain’s MCP server support, signaling compatibility with broader agent harnesses.
- company-brain and client-pods: Terms associated with the broader organizational/agent architecture built around GBrain.
- github: The distribution point for GBrain and its eval artifacts.
- karpathy-style-knowledge-wiki and git-history / dry / gstack / zeroentropy / voyage-ai / openai / opus / hyojun_at: Related ecosystem references that connect to the broader tooling, knowledge, retrieval, and agent workflows discussed around GBrain in the newsletter.
Frequently asked questions
What is GBrain?
A product or system referenced as a retrieval and latency data task target. The newsletter uses it as an example of where strong evaluations around retrieval and cost matter.
What should AI product managers know about GBrain?
GBrain is positioned as an MIT-licensed open-source retrieval and memory system for AI agents and company-brain workflows. Newsletter coverage ties GBrain closely to retrieval quality, latency, and cost-per-compute evaluation discipline. It was described as state-of-the-art for agentic retrieval without LLM query rewriting, with public eval references in gbrain-evals.
Explore this topic
Newsletter Mentions (18)
“#2 𝕏 Garry Tan shared his gbrain-evals repository, which includes 40+ categories that can each be auto-researched and auto-improved. He said clear evaluations for retrieval, latency, and cost per compute are useful for retrieval and latency data tasks like those performed by GBrain.”
#2 𝕏 Garry Tan shared his gbrain-evals repository, which includes 40+ categories that can each be auto-researched and auto-improved. He said clear evaluations for retrieval, latency, and cost per compute are useful for retrieval and latency data tasks like those performed by GBrain.
“Garry Tan shared the gbrain GitHub repository, describing it as offering a private GitHub repo with 70 of his “proven skills” and the beginnings of a “Karpathy-style knowledge wiki.””
GenAI PM Daily August 18, 2026 GenAI PM Daily 🎧 Listen to this brief 3 min listen Today's top 20 insights for PM Builders, ranked by relevance from X, YouTube, LinkedIn, and Blogs. Cursor releases Origin, its integrated code hosting platform #1 𝕏 Cursor released Origin, its code hosting platform, with deep Cursor integration and repository syncing from GitHub. Cursor describes Origin as fast and easy to use. Also covered by: @Cursor , @Guillermo Rauch #2 𝕏 Philipp Schmid demonstrated Gemini 3.7 Flash using his Android emulator via ADB for the task “Play 1 round of Wordle.” He said its latency and visual reasoning make it exceptionally good for multimodal agentic use cases such as mobile control and Computer Use. #3 𝕏 LlamaIndex 🦙 recapped ExtractBench, which requires both extracted values and citations to be correct and evaluates word-level boxes at IoU 0.5. LlamaExtract Agentic Plus led with 84.9% page-level and 46.4% word-level results, achieving 87.1% on long documents where other systems scored zero. #4 𝕏 Garry Tan shared the gbrain GitHub repository, describing it as offering a private GitHub repo with 70 of his “proven skills” and the beginnings of a “Karpathy-style knowledge wiki.” The material is free, open source, and MIT-licensed, with full documentation in the README.
“Garry Tan released GBrain v0.45.6.0 with 17 new brain skills, hardened through his personal OpenClaw agent using hundreds of thousands of markdown files.”
#15 𝕏 Garry Tan released GBrain v0.45.6.0 with 17 new brain skills, hardened through his personal OpenClaw agent using hundreds of thousands of markdown files. GBrain now works with Codex and Claude Code.
“Garry Tan demonstrates that GBrain is now state-of-the-art for agentic retrieval without any LLM rewriting, with full evaluation results in the gbrain-evals GitHub repo.”
GenAI PM Daily July 28, 2026. GBrain is discussed as a retrieval breakthrough with supporting evaluation artifacts.
“Garry Tan launched GBrain, a free open-source retrieval library optimized for Hermes Agent and OpenClaw (with Codex and Claude Code support) that powers his personal company brain and AI.”
Presented as an open-source retrieval layer for agentic knowledge access and personal AI memory.
“Garry Tan introduced GBrain, an MIT-licensed open-source retrieval layer for AI agents—acting like a Postgres-style, just-in-time “3-book librarian”—and it’s already in use by thousands of developers.”
#9 𝕏 Garry Tan introduced GBrain, an MIT-licensed open-source retrieval layer for AI agents—acting like a Postgres-style, just-in-time “3-book librarian”—and it’s already in use by thousands of developers.
“Summary: Garry Tan unveiled GBrain, a modular “company brain” framework that structures work via scoped AI agents organized into client pods.”
#12 𝕏 Summary: Garry Tan unveiled GBrain, a modular “company brain” framework that structures work via scoped AI agents organized into client pods. This detailed agent company architecture standardizes workflows and scales cross-functional collaboration.
“Garry Tan open-sourced GBrain (MIT-licensed) on GitHub and outlines a 30-minute setup using his 350k-page markdown LLM wiki plus an OpenClaw/Hermes agent that automates most tasks.”
#6 𝕏 Garry Tan open-sourced GBrain (MIT-licensed) on GitHub and outlines a 30-minute setup using his 350k-page markdown LLM wiki plus an OpenClaw/Hermes agent that automates most tasks.
“Garry Tan launched GBrain, an MIT-licensed, state-of-the-art retrieval engine for agents—built for OpenClaw and Hermes but with full MCP server support to plug into almost any agent harness.”
#6 𝕏 Garry Tan launched GBrain, an MIT-licensed, state-of-the-art retrieval engine for agents—built for OpenClaw and Hermes but with full MCP server support to plug into almost any agent harness. #10 𝕏 Garry Tan rolled out the latest GBrain update, which adds synthesized answers to your queries instead of just basic retrieval. An A/B test of GBrain Search vs. GBrain Think shows it improving in accuracy every single day.
“Garry Tan released GBrain—an MIT-licensed OSS retrieval and memory system that achieves SOTA on LongMemEval, beating all known open-source repos by over 1% without any LLM query rewriting.”
#12 𝕏 Garry Tan released GBrain—an MIT-licensed OSS retrieval and memory system that achieves SOTA on LongMemEval, beating all known open-source repos by over 1% without any LLM query rewriting. He’s already running it on his own 100k-page OpenClaw/Hermes Agent brain.
Related
Anthropic’s coding agent experience referenced as one of the targets for reusable skills. It is part of the multi-agent workflow ecosystem described in the newsletter.
An AI research and product company behind models, tools, and repositories. In this newsletter it is referenced through the openai/math repository.
An AI coding agent referenced as a target platform for reusable software-engineering skills. It is listed alongside other coding assistants in the workflow ecosystem.
Investor and startup leader mentioned sharing eval tooling and operational advice for agents and CI workflows. The newsletter links him to repositories and experimental modes used for agentic engineering.
A partner company listed in the Sign in with ChatGPT launch set. The newsletter provides no additional context beyond the integration mention.
A protocol used for connecting tools and agents through standardized interfaces. In this newsletter it appears in the context of servers and tools for managing agent context and integrations.
A product or integration layer that now includes Perplexity Search. The newsletter only mentions it as the destination where Perplexity Search is embedded.
The code-hosting and collaboration platform used for pull requests and repository workflows. Here it is part of Warp’s automated engineering loop.
A pattern for using retrieved context in AI applications. It appears here alongside tools for retrieval, compression, and evaluation.
A model used in the newsletter as a reasoning and execution engine for product experimentation. It is described as generating daily A/B test ideas and implementing winners for a mobile game economy.
An AI agent environment or product that can host models and persona features. In this newsletter it appears both as a place where Qwen3.8-Max is available and as a tool with a /personality feature.
Stay updated on GBrain
Get curated AI PM insights delivered daily — covering this and 1,000+ other sources.
Subscribe Free