Replicate
Replicate lets you run machine learning models in the cloud with a simple API. Thousands of open-source models are available, and you can run your own custom models at scale. Run image generation, language models, audio synthesis, video generation, and more with a few lines of code. Replicate makes AI accessible to every software engineer.
Real signal across most facets with visible, nameable gaps — the contract exists but is thin, or the portal is good while governance and commercial terms are absent.
API Evangelist profiles Replicate the way a machine reads it — 85 machine-readable artifacts across 1 API, pulled from the provider's own public surface and indexed so a developer, an analyst, or an AI agent can evaluate it against every other provider on the network.
Every provider in the network is reduced to the same set of machine-readable artifacts — OpenAPI contracts, event specifications, GraphQL schemas, runnable collections, pricing and rate-limit signals, security posture, OAuth scopes, and the agent surfaces (MCP servers and skills) that let software drive the API on its own. We profile them because the interface is the part of a company you can actually inspect: it is a truer signal of what a provider does than any marketing page. From those artifacts we compute the Kin Score — Replicate scores 53.9/100 (developing), with a separate agent-readiness read of 30/100 (agent ready). The full breakdown is below, followed by every artifact we hold — each card links through to its machine-readable definition on apis.io.
Kin Score
This is the API Evangelist rating — a single, repeatable read computed from the artifacts on this page. Green fill is points earned; the red track is points possible, so every bar shows earned-versus-possible at a glance. Every facet and dimension name is a link: it opens that measurement's page on APIs.io, where the rating runs across the whole catalog — the exact checks that feed it, how every profiled provider distributes on it, and who is at the top of it.
Kin Score
How this is scored →scored 2026-09-08 · rubric v0.20.0
Put this on your own site. The badge is drawn live from Replicate's current Kin Score — paste it once and it updates itself every time the score is recomputed. It follows your visitor's light or dark setting, and it links back here so anyone who sees it can read the full breakdown.
<!-- Kin Score · API Evangelist -->
<a href="https://providers.apievangelist.com/providers/replicate/"
title="Replicate on API Evangelist — API profile and Kin Score">
<img src="https://apis.io/badge/replicate.svg"
alt="Replicate Kin Score — API readiness rating by API Evangelist" width="150" height="150" loading="lazy">
</a>[](https://providers.apievangelist.com/providers/replicate/)<!-- Kin Score · API Evangelist -->
<a href="https://providers.apievangelist.com/providers/replicate/"
title="Replicate on API Evangelist — API profile and Kin Score">
<img src="https://apis.io/badge/replicate/card.svg"
alt="Replicate Kin Score — API readiness rating by API Evangelist" width="340" height="120" loading="lazy">
</a>More shapes, themes and sizes → · Score as JSON · How badges work
How we profile Replicate
Each block below is one kind of artifact we hold for Replicate. For each we say what it is and why it earns a place in the profile, then list every one we've indexed — capped at two rows, scroll within the panel for the rest.
APIs 11
Each API is captured as its own OpenAPI definition — every operation, parameter, and response. This is the single most useful machine-readable description of what an API does, and it's what lets us score, lint, mock, and generate against it without asking the provider for anything.
Individual APIs this provider publishes, each with its own machine-readable definition.
Replicate Accounts API
The Accounts API from Replicate — 1 operation(s) for accounts.
Replicate Cancel API
The Cancel API from Replicate — 2 operation(s) for cancel.
Replicate Collections API
The Collections API from Replicate — 2 operation(s) for collections.
Replicate Deployments API
The Deployments API from Replicate — 3 operation(s) for deployments.
Replicate Hardware API
The Hardware API from Replicate — 1 operation(s) for hardware.
Replicate Model API
The Model API from Replicate — 5 operation(s) for model.
Replicate Models API
The Models API from Replicate — 1 operation(s) for models.
Replicate Predictions API
The Predictions API from Replicate — 5 operation(s) for predictions.
Replicate Secrets API
The Secrets API from Replicate — 1 operation(s) for secrets.
Replicate Training API
The Training API from Replicate — 2 operation(s) for training.
Replicate Trainings API
The Trainings API from Replicate — 1 operation(s) for trainings.
Scroll within the panel for all 11 ·
Postman Collections 1
A runnable collection turns the contract into something a developer can execute in seconds. We profile them because the fastest way to trust an API is to make a real call against it.
Ready-to-run Postman collections for exercising this provider's APIs.
Replicate
POSTMANOpen Collections 18
Open, tool-agnostic collections carry the same runnable value as Postman without locking you to one client — the portable, forkable form of the same exercise.
Open, tool-agnostic API collections (OpenAPI-derived and Bruno).
API Collection
OPEN COLLECTIONReplicate Accounts API
OPEN COLLECTIONReplicate Accounts Cancel API
OPEN COLLECTIONReplicate Accounts Collections API
OPEN COLLECTIONReplicate Accounts Deployments API
OPEN COLLECTIONReplicate Accounts Hardware API
OPEN COLLECTIONReplicate Accounts Model API
OPEN COLLECTIONReplicate Accounts Models API
OPEN COLLECTIONReplicate Accounts Name API
OPEN COLLECTIONReplicate Accounts Owner API
OPEN COLLECTIONReplicate Accounts Predictions API
OPEN COLLECTIONReplicate Accounts Secrets API
OPEN COLLECTIONReplicate Accounts Slug API
OPEN COLLECTIONReplicate Accounts Training API
OPEN COLLECTIONReplicate Accounts Trainings API
OPEN COLLECTIONReplicate Accounts Version API
OPEN COLLECTIONReplicate Accounts Webhooks API
OPEN COLLECTIONReplicate
OPEN COLLECTIONScroll within the panel for all 18 ·
Arazzo Workflows 10
Real integrations are rarely a single call. Arazzo describes the multi-step sequences — auth, then create, then confirm — so both a human and an agent can follow the choreography, not just the endpoints.
Multi-step API workflows described with the Arazzo specification.
Replicate Pick a Model from a Collection and Predict
Read a curated collection, confirm a chosen model, run its latest version, and poll the prediction.
ARAZZOReplicate Create a Deployment and Run a Prediction Through It
Pick hardware, create a deployment for a model version, then run a prediction via the deployment.
ARAZZOReplicate Resolve Latest Version and Predict
Look up a model, pick its latest version, run a prediction, and poll to completion.
ARAZZOReplicate Run an Official Model and Poll
Run a prediction against an official model by name, then poll until complete.
ARAZZOReplicate Create Prediction and Poll Until Complete
Run a model version, then poll the prediction until it reaches a terminal state.
ARAZZOReplicate Run a Prediction with Bounded Wait and Cancel
Create a prediction, poll a bounded number of times, and cancel it if it has not finished.
ARAZZOReplicate Scale a Deployment and Run a Prediction
Read a deployment, update its version and instance bounds, then run a prediction through it.
ARAZZOReplicate Search for a Model and Run a Prediction
Search public models by query, run a prediction on the top match's latest version, and poll it.
ARAZZOReplicate Start a Training and Poll Until Complete
Start a fine-tuning run from a base model version, then poll until the training finishes.
ARAZZOReplicate Fetch Webhook Secret and Run a Webhook Prediction
Retrieve the default webhook signing secret, then create a prediction that posts to a webhook.
ARAZZOScroll within the panel for all 10 ·
MCP Servers 1
Model Context Protocol servers expose these APIs directly to AI agents. We profile them because agent-native access is the fastest-growing way this provider's capabilities actually get used.
Model Context Protocol servers that expose these APIs to AI agents.
MCP Server
MCP SERVERAgent Skills 7
An agent skill packages the how-to for driving these APIs from an assistant — the prompts, the sequence, the guardrails — so the knowledge to use the API travels with it.
Packaged agent skills for driving this provider's APIs from an AI assistant.
build-models
AGENT SKILLcompare-models
AGENT SKILLfind-models
AGENT SKILLprompt-images
AGENT SKILLprompt-videos
AGENT SKILLpublish-models
AGENT SKILLrun-models
AGENT SKILLScroll within the panel for all 7 ·
GraphQL 1
Where a provider ships GraphQL, the schema is the contract. We profile it alongside the REST surface so the whole interface is legible in one place.
GraphQL schemas published by this provider.
Replicate GraphQL
Replicate does not currently expose a public GraphQL endpoint. The platform's primary API is a REST HTTP API available at `https://api.replicate.com/v1`. This GraphQL schema is ...
GRAPHQLPricing Plans 1
Pricing is part of the interface. Machine-readable plans tell you what a tier costs and includes before you commit — one of the things the Kin Score reads for access clarity — renamed from commercial clarity in rubric 0.12, because a free statutory interface has access terms and no commercial ones.
Published pricing tiers and plan structures.
Rate Limits 1
Rate limits are the difference between a demo that works and a production integration that doesn't fall over. Publishing them is an operational-transparency signal — and a hard requirement for any agent that plans its own throughput.
Documented rate limits and quota policies.
Replicate Rate Limits
RATE LIMITSFinOps 1
Cost, billing, and metering signals let a buyer model the financial operations of an API before it's live. We profile them for the same reason we profile pricing: the money is part of the contract.
Cost, billing, and metering signals for API financial operations.
Replicate Finops
FINOPSFeatures 16
The notable capabilities this provider advertises, captured as structured features so they can be searched and compared instead of read one landing page at a time.
Notable capabilities this provider offers.
Scroll within the panel for all 16 ·
Event Specifications 1
Not every API is request/response. AsyncAPI describes the event-driven and streaming side — the webhooks and channels — so the asynchronous half of the interface is documented the same way the synchronous half is.
AsyncAPI definitions for this provider's event-driven and streaming APIs.
Replicate Streaming and Webhooks API
AsyncAPI definition for Replicate's event-driven surfaces: - Server-Sent Events (SSE) stream returned for predictions where the model supports streaming output. The stream URL i...
ASYNCAPISemantic Vocabularies 1
JSON-LD contexts give the data shared meaning across APIs. We profile them because semantics are what let a machine reconcile 'customer' here with 'customer' somewhere else.
JSON-LD contexts and semantic vocabularies used across these APIs.
Replicate Context
JSON-LDSpectral Rules 3
Governance rulesets we run against this provider's specs — the automated checks behind parts of the score. Profiling them makes the quality bar explicit and re-runnable, not a matter of opinion.
JSON Schema 5
Standalone JSON Schema definitions describe the data models behind the API. We profile them so the shapes are validatable on their own — useful long after a single request is forgotten.
Standalone JSON Schema definitions for this provider's data models.
Replicate Model
JSON SCHEMAReplicate Prediction
JSON SCHEMAschemas_prediction_request
JSON SCHEMAschemas_training_request
JSON SCHEMAschemas_version_prediction_request
JSON SCHEMAJSON Structure 2
JSON Structure captures the data shapes in a form built for tooling — a complement to JSON Schema that keeps the model machine-legible.
JSON Structure definitions describing this provider's data shapes.
Replicate Prediction Structure
JSON STRUCTUREReplicate Structure
JSON STRUCTUREExamples 2
Real request and response payloads are what turn a spec from abstract into obvious — and they're one of the twelve things an agent needs to call an API correctly on the first try.
Example request and response payloads for these APIs.
Security Posture 2
Authentication, domain security, vulnerability disclosure, and trust-center signals — the evidence that a provider takes security seriously enough to document it. We profile it because you can't govern what you can't see.
Authentication, domain security, vulnerability disclosure, and trust-center signals.
Agentic Access 1
An x-agentic-access contract marks which operations are safe for an agent to run on its own and which need a human in the loop. It is the difference between an API an agent can use and one it can use safely.
Recommended x-agentic-access execution contracts for AI agents.
Resources
Every other property we hold for Replicate — documentation, portals, status pages, policies, and corporate surface — grouped by the job it does, following the integrator's arc from getting started to running in production.
Get Started 2
Portal, sign-up, and the first successful call
Documentation 1
Reference material describing how the API behaves
Agent Surfaces 4
MCP servers, agent skills, and machine-readable catalogs
Design & Contract 10
Pagination, idempotency, versioning, errors, and events
Scroll within the panel for all 10 ·
Build 7
SDKs, sample code, and the tooling you integrate with
Scroll within the panel for all 7 ·
Access & Security 2
Authentication, authorization, and security posture
Operate 2
Status, limits, changes, and where to get help
Commercial 3
Pricing, plans, and the legal terms of use
Company 3
The organization behind the API
Other 2
Properties that don't map to a standard resource type
← All providers · Data indexed from github.com/api-evangelist/replicate · machine-readable index on apis.io
This is an independent, third-party profile of Replicate, published by API Evangelist. We do not operate, host, resell, or support these APIs, and we are not affiliated with or endorsed by the company unless stated above. Everything here is built from publicly available information — the company's own site, developer portal, documentation, public repositories, and the specifications it publishes for public use. Nothing is obtained by breaching a system, defeating an access control, or using credentials.
The Kin Score and Agent Readiness rating are independently calculated assessments of a company's public API artifacts, scored against a published rubric. They are not certifications, endorsements, security assessments, or audits.
Corrections, re-scores, and removal are free — no partnership or purchase required, and you do not need to justify the request. A removed company is recorded as unrated, never scored zero for having asked. Acknowledgement within one business day; removal within two.
info@apievangelist.com
·
Read the full data-sourcing policy →
On a security or compliance team? Put security in the subject line and
you will get a person, not a form — we will tell you exactly which public URLs this profile was built from.