# MemVerge

> MemVerge builds memory-centric AI infrastructure. The company's premise is
> that today's AI is model-centric — you orbit the model — and that it should be
> memory-centric instead: your context, history and identity stay put while
> models become tools that rotate around you.

MemVerge products span personal AI memory (MemoryBox), enterprise GPU
orchestration and transparent checkpointing (Memory Machine AI), CXL memory
expansion and sharing (Memory Machine X), and the open-source MemMachine
project.

## Products

- [MemoryBox](https://www.memorybox.ai/): a personal AI memory vault. Context follows you across ChatGPT, Claude, Gemini and any other assistant, stored locally on your own machine. Free beta for macOS and Windows.
- [MemMachine](https://memmachine.ai/): the open-source memory layer underneath MemoryBox.
- [Memory Machine AI](https://memverge.ai/product/memory-machine-ai): GPU orchestration and AI workload automation for enterprises, on prem and in the cloud.
- [Transparent Checkpointing](https://memverge.ai/product/snapshot): checkpoint and restore running AI jobs with no changes to the application — survive GPU failure, use spot instances safely, get fast cold starts.
- [Memory Machine X for CXL](https://memverge.ai/product/memory-machine-cxl): CXL memory expansion, tiering and fabric-attached memory for big-memory data centers.

## Key pages

- [Home](https://memverge.ai/)
- [About MemVerge](https://memverge.ai/company/about)
- [Leadership](https://memverge.ai/company/leadership)
- [Contact Us](https://memverge.ai/company/contact)
- [Blog](https://memverge.ai/blog): engineering deep-dives and ideas from the MemVerge team.
- [News](https://memverge.ai/news): announcements, press coverage and events.
- [Privacy Policy](https://memverge.ai/privacy)
- [Terms of Service](https://memverge.ai/terms)

## Use cases

- [Maximize AI workload throughput with GPU fractionalization](https://memverge.ai/product/use-case-maximize-ai-workload-throughput-and-cost-efficiency-with-gpu-fractionalization)
- [Integrated GPU resource management for enterprise AI](https://memverge.ai/product/use-case-integrated-gpu-resource-management-for-efficient-productive-enterprise-ai-environments)
- [Never lose progress during GPU failure or node maintenance](https://memverge.ai/product/use-case-never-lose-progress-during-gpu-failure-node-maintenance)
- [Run stateful AI jobs safely on low-cost cloud spot instances](https://memverge.ai/product/use-case-run-stateful-ai-jobs-safely-on-low-cost-cloud-spot-instances)
- [Fast cold starts and hot restarts](https://memverge.ai/product/use-case-fast-cold-starts-and-hot-restarts)
- [CXL: slash memory costs and expand capacity](https://memverge.ai/product/cxl-use-case-slash-memory-costs-and-expand-capacity)
- [CXL for vector databases](https://memverge.ai/product/cxl-use-case-vector-database)
- [CXL for MySQL](https://memverge.ai/product/cxl-use-case-mysql-database)
- [CXL for Ray](https://memverge.ai/product/cxl-use-case-ray)

## Documentation

- MemoryBox docs: https://docs.memorybox.ai/
- MemMachine docs: https://docs.memmachine.ai/

## Other languages

- 简体中文: https://memverge.ai/zh-CN