Which LLM is the cheapest for our workload?

The best-for pages rank the live catalogue per task by the index that measures it, and the router picks the cheapest model that meets your quality rule for each request.

How much can we save on our LLM bill?

On the public benchmark the router saved 89% against sending everything to the reference model. Import last month's provider bill on the bill page and the estimate is computed on your own traffic before you change a line of code.

How often do model prices change?

The catalogue is refreshed from the vendors' price lists; each model page shows the date of its data.

90 days of the enterprise, pro and team plans free, no card needed. See the plans

OpenSmartRoute

Rankings

What the marketplace installs and rates.

Published listings ranked per kind: the most installed, the best rated (a Bayesian average that a single five-star review cannot top) and the newest. Every row links to the listing; publishers appear by name.

Ranked listingsmarketplace index · all sources: 38.7K
Installsinto workspaces here · all time: 0
Reviewsby people who installed · all time: 0

All kinds Agents8 Skills14.7K Personas3.3K Prompts4 Templates37 Tools20.1K Models482Most installed Best rated Newest

Prompts ranked by most installed
#	Listing	Publisher	Installs	Rating	Version	Updated

Raw numbers: /api/v1/rankings/listings and the kinds at /api/v1/rankings/listings/kinds. Browse everything on the marketplace.

Rated by the people who installed it

A listing climbs this board by being installed and kept, not by being new.

Installs count once per workspace, ratings are averaged the Bayesian way so a single five-star review cannot top the board, and publishers appear by name beside their work. Open any row to read what it does, what it costs and who stands behind it before you add it to a router.

Photo: RDNE Stock project

For the business

What it means for which models you pay for.

Catalogue of every vendor
The right model, not the famous one
Price, context, abilities and benchmark indexes side by side, with the model that fits each task named.
The rankings
89% less
Lower cost per request
On the 150-prompt benchmark the router spent 89% less than sending every prompt to the reference model, and the judged quality went up, not down.
Read the benchmark
Per million tokens, every model
Prices you can compare

Install one, or publish your own.

Every listing routes like a model: declare it as a target and the router picks it when it is the best answer. Publishers list for free and are rated by the people who install.

Browse the marketplace Publish a listing

1	RFC / design-doc drafter Prompt Expands a rough idea into a review-ready design doc: context, options, trade-offs, decision.	OpenSmartRouteverified	(0)	1.0.0	Oct 16, 2026
2	Bug report triage Prompt Turns a messy bug report into severity, steps to reproduce, expected vs actual and next action.	OpenSmartRouteverified	(0)	1.0.0	Oct 16, 2026
3	Safe SQL analyst Prompt Writes one read-only SQL query for a question, explains it, and refuses anything that writes.	OpenSmartRouteverified	(0)	1.0.0	Oct 16, 2026
4	Meeting notes to decisions Prompt Turns raw meeting notes into decisions, owners, due dates and open questions.	OpenSmartRouteverified	(0)	1.0.0	Oct 16, 2026

What the marketplace installs and rates.

A listing climbs this board by being installed and kept, not by being new.

What it means for which models you pay for.

The right model, not the famous one

Lower cost per request

Prices you can compare

Install one, or publish your own.