
supercomputer
From $0.47 per useActively maintained
not read yet
Use
- For
- Developers and scientists running GPU jobs from coding agents or AI assistants
- Price
- From $0.47 per usepricing page, read Oct 27, 2026
- Activity
- Actively maintainedsite updated 6 days ago
- How it runs
- CLI
- Used for
- Launch GPU training jobs from a coding agent
- Run an overnight metric-driven experiment loop on one GPU
- Fine-tune a robot policy on a single GPU
- Has an MCP server (1 in the registry)public registry →
- Publishes llms.txt, a file written for AI assistantssupercomputer.sh ↗
- Privacy policy updated2026-10-13privacy policy ↗
These are supercomputer's own statements, read from its pages. We did not audit them. A line that is missing was not found on the pages we read; it does not mean the practice is absent.
- MCP serversh.supercomputer/supercomputerverified in the MCP registry
- Stateanswers, requires sign-inchecked 2026-10-27
- Address to connectsupercomputer.sh/mcp
- MCP docssupercomputer.sh/docs/guides/mcpthe vendor's page ↗
This card as JSON, for your agent →
Each line is something a program found when it went looking: the registry entry, the server's own answer, a spec that parses. A missing line means we did not find it, not that it does not exist.
Figure and bar: visits/mo, estimate · year: online since
- #1

Together AI
Runs open-source language models with serverless inference, batch processing, and GPU clusters.
32.7kActive2022 - #2

Modal
Runs inference, training, and batch processing workloads on GPUs with instant autoscaling and sub-second cold starts.
52.1kActiveFrom $250/mo - #3

GMI CloudMCP · sign-in
Runs AI model inference and agents on serverless or dedicated GPU infrastructure with automatic scaling.
19.7kActiveFrom $2.60 per use - #4

Outlier
Runs large language models locally on Apple Silicon Macs for coding, file editing, and research without sending data to servers.
16ActiveFrom $249 once - #5

KourierMCP · open
Serves open-weight language models through an OpenAI-compatible API endpoint running on the company's own GPU clusters.
2ActiveFrom $20/mo
Not it? Say what you need
betaThree tools from the Atlas for the job you describe, each with what fits and what does not.
No notes yet. What do you use supercomputer for?
Sign in to leave a noteWhat is supercomputer?
Rents GPU and CPU sessions for agents and developers that keep code and files between runs and bill only while running. It is listed in the Atlas under LLM operations.
Who is supercomputer for?
Developers and scientists running GPU jobs from coding agents or AI assistants
How much does supercomputer cost?
Paid plans start at $0.47 per use, per supercomputer's pricing page on 2026-10-27.
Is supercomputer free?
No. Its pricing page showed no free plan on 2026-10-27.
Do people use supercomputer?
Little sign of use.
Is supercomputer still maintained?
Actively maintained: site updated 6 days ago.
What are alternatives to supercomputer?
In LLM operations, by use: Together AI, Modal, GMI Cloud, Outlier, Kourier.
How we read a tool
No votes, no reviews, no vendor claims. Every week a crawler reads each tool's own site and a few public registries, and the words on the card are bands over what it read.
- Use: visits to the site (estimated), installs from npm and PyPI, presence in Chrome's usage report.
- Activity: the newest release on GitHub, npm or PyPI; the newest dated page on the site; open roles on a public jobs board.
- Price: the vendor's own pricing page, read with the date. A figure is printed only when it is on that page.
- The line and the use cases are written by us from the homepage, one tool at a time, and refused when they repeat the vendor's marketing.
How AI assistants read each site — the reading vendors ask us about — is on each tool's own visibility page.
How we read every tool → · By INITE Atlas Desk
Cookies
Before we count this visit
Analytics and ad measurement tell us which pages and which of our ads do their job: Google Analytics and Google Ads, the Meta pixel, Yandex Metrica, and one first-party id of our own. Nothing is stored until you answer, and declining changes nothing about the site.