Who should use RiftStack?

RiftStack is an open-source project built for AI researchers and inference engineers.

Does RiftStack have a free plan?

Yes — RiftStack is open source under the Apache-2.0 license and free to use.

What platforms does RiftStack run on?

RiftStack runs on the web and API. It can also be self-hosted.

Is RiftStack still active?

The GitHub repository shows 332 commits in the last 90 days.

What projects are similar to RiftStack?

Similar projects tracked by PulseGate include Cliprise, chi-ai, and AI DocuChat.

Who develops RiftStack?

RiftStack is developed by RiftStack.

How long has RiftStack been around?

RiftStack first shipped in 2025.

Back to the index

RiftStack

Name: RiftStack
Author: RiftStack

riftstack.aiInfrastructure

RiftStack researches and builds tools for inference optimization, focusing on compiler technology and serving techniques. Its open-source ML compiler Emmy transforms Torch IR through multiple representations down to optimized CUDA kernels. The lab publishes tutorials and benchmarks on GPU performance, memory virtualization for long-context inference, and kernel optimizations for RTX GPUs.

Open SourceApache-2.0WebAPISelf-hosted

Visit riftstack.ai

67stars

8forks

5features

2025since

Overview

5 features

In the AI & ML space, RiftStack takes a focused approach. It focuses on running large language models efficiently and cost-effectively on consumer and commodity GPUs rather than only high-end datacenter hardware. RiftStack is an open-source project aimed at AI researchers and inference engineers. RiftStack is open source under the Apache-2.0 license. RiftStack is available on the web and API, and it can be self-hosted.

RiftStack builds and maintains RiftStack, and it first shipped in 2025. The project is developed in the open on GitHub with 67 stars and 332 commits in the last 90 days. Among its 5 catalogued features are ML Compiler, Kernel Optimization, and GPU Matmul Tuning.

Summary written by a language model from the project’s public pages.

ML Compiler
Kernel Optimization
GPU Matmul Tuning
KV Cache Virtualization
Inference Benchmarking

Built with & integrations

Framework

Next.js

Hosting

Vercel

AI providers

local_oss

Runs on

BrowserAPI-onlySelf-hosted

Detected from

Next.js: x-powered-by header · /_next/static/ in the HTML
Vercel: x-vercel-id header · x-vercel-cache header

Trust & compliance

License

Apache-2.0

Verified signals

HTTPSPrivacy Policy Terms of Service Open SourceFree tierGitHub · ★ 67Active maintenance

Legal

Indexing history

What PulseGate has recorded for this listing

Indexed28 Jul · 14:54 UTC
Exploring FlashAttention-3/4 optimizations on RTX GPUs verified against its public source
Source: PulseGate · Open

Frequently asked questions about RiftStack

What does RiftStack do?: RiftStack focuses on running large language models efficiently and cost-effectively on consumer and commodity GPUs rather than only high-end datacenter hardware. It is catalogued under AI & ML on PulseGate.
Who should use RiftStack?: RiftStack is an open-source project built for AI researchers and inference engineers.
Does RiftStack have a free plan?: Yes — RiftStack is open source under the Apache-2.0 license and free to use.
What platforms does RiftStack run on?: RiftStack runs on the web and API. It can also be self-hosted.
Is RiftStack still active?: The GitHub repository shows 332 commits in the last 90 days.
What projects are similar to RiftStack?: Similar projects tracked by PulseGate include Cliprise, chi-ai, and AI DocuChat.Cliprise chi-ai AI DocuChat
Who develops RiftStack?: RiftStack is developed by RiftStack.
How long has RiftStack been around?: RiftStack first shipped in 2025.

At a glance

Pricing: Open Source
Platforms: Web
Languages: English
Open source: Yes · ★ 67
License: Apache-2.0
Built for: AI researchers and inference engineers
Model: Open source
Solves: Running large language models efficiently and cost-effectively on consumer and commodity GPUs rather than only high-end datacenter hardware.

Registered as

GitHub: cloudrift-ai/emmy

Developer

Team

GitHub

Open source

View on GitHub

Stars: 67
Forks: 8
Open issues: 7
Last commit: 28 Jul 2026
Commits 90d: 332
Contributors: 7
Authorship: Team
Default branch: main
Latest release: v0.2.0 · 10 Jul 2026

Live coverage

Identity confidence: Medium · 72.8
Indexed: 28 Jul 2026
Lifecycle: Alive
Last seen: 28 Jul 2026

Identity audit (13)

Slug: exploring-flashattention-3-4-optimizations-on-rtx-gpus-riftstack-ai
Lifecycle state recorded: 28 Jul 2026
Verification state: Indexed for public listing
Listing state: Listed: yes
Index status: Included in index
Latest evidence snapshot: 28 Jul 2026
Timeline basis: Indexed-at chronology. This listing's first-seen date was written by the catalog backfill, not observed here, so it is not treated as a sighting.
Name from: Derived from the project's own page and URL.
Category from: Assigned by a language model.
Summary from: Written by a language model from public pages.
Languages from: Detected by a language model, checked against the page's own declaration.
Last updated: 28 Jul 2026
Canonical URL: https://riftstack.ai/

Ship this? Send a correction — no account, and you get a link to follow it.

Also in AI & ML

Same category — not a similarity match