Replicate
- Company typePrivate
- Founded2023
- HeadquartersSan Francisco, United States
- Headcount11–50
- GTM typeB2B
- OfferingSoftware
Replicate firmographics
Firmographics- Name
- Replicate
- Legal name
- Replicate, LLC
- Website
- https://replicate.com
- Company type
- Private
- Founded year
- 2023
- Operating status
- Acquired
- Headcount range
- 11–50 employees
- Ownership category
- akta.pro rank
Replicate industry classification
Industry- Product category
- AI Inference Platform
- NAICS
- Computing Infrastructure Providers, Data Processing, Web Hosting, and Related Services (5182), Software Publishers (51321)
- SIC
- Services-Prepackaged Software (7372), Services-Computer Programming, Data Processing, Etc. (7370)
- akta.pro primary industry
- Model Deployment, Serving & Inference Platforms (HDAAABAF)
- akta.pro secondary industries
- Model Hosting, Serving & Inference Platforms (HDAAACAB), Model Serving, Inference & Deployment Platforms (APIs, Edge/On-Prem) (HDAEANAC), End-to-End Enterprise AI Platforms (MLOps & Model Lifecycle Management) (HDAEANAA)
Keywords
Where Replicate is headquartered
LocationHeadquarters
- HQ city
- San Francisco
- HQ country
- United States
- HQ region
- North America
Offices1 record
Markets served
Replicate business model
Business model- GTM type
- B2B
- Offering type
- Software
- Cost components
- Infrastructure, Technology or R&D, Personnel, Operations, Marketing or Sales
Revenue model
- Compute/Usage-Based Billing: Replicate bills customers for compute time used when running models. Customers pay per second of GPU time depending on hardware type selected (CPU, T4, L40S, A100). No charge when not actively running. Setup and idle time is free for public models.
- Prepaid Credit: Customers can prepay credit which is deducted as they use the service. Credits expire after 12 months if unused. Auto-reload feature available.
- Payment in Arrears: Enterprise customers may be billed monthly in arrears for usage from the previous month. Supports larger organizations with established billing.
Pricing tiers
| Model | Billing | Price |
|---|---|---|
| Usage-based | Pay-as-you-go | CPU compute |
| Usage-based | Pay-as-you-go | Nvidia T4 GPU |
| Usage-based | Pay-as-you-go | Nvidia L40S GPU |
| Usage-based | Pay-as-you-go | 2x Nvidia L40S GPU |
| Usage-based | Pay-as-you-go | Nvidia A100 (80GB) GPU |
| Usage-based | Pay-as-you-go | 8x Nvidia A100 (80GB) GPU |
| Usage-based | Pay-as-you-go | Video Generation - 720p |
| Usage-based | Pay-as-you-go | Video Generation - 1080p |
| Freemium | Pay-as-you-go | Free Tier |
| Hybrid | Multi-year contract | Enterprise Plans |
Go-to-market motion2 records
Distribution channels4 records
Marketing channels5 records
Replicate product offering
Product offeringCore offering
Replicate is a cloud-based AI model deployment platform that lets developers run, fine-tune, and deploy over 50,000 open-source and proprietary machine learning models via a single API. The platform abstracts GPU infrastructure management, auto-scaling, cold boots, and versioning, allowing software developers to integrate AI capabilities into their applications with one line of code across image, video, audio, and text modalities. Supporting tooling includes Cog, an open-source Docker-based ML model packaging utility, and LoRA-based fine-tuning for image models such as SDXL and FLUX.
Product overview
Replicate is a cloud-based AI model hosting and deployment platform acquired by Cloudflare in early 2026. The platform provides a unified API and web interface for running, fine-tuning, and deploying 50,000+ machine learning models from providers including Meta, Google, Microsoft, OpenAI, Anthropic, and Black Forest Labs. The core Replicate Platform enables one-line model execution with automatic GPU scaling and pay-per-use pricing. Key supporting products include Cog (open-source Docker-based model packaging tool), LoRA Training and Inference (fast fine-tuning for Stable Diffusion), Deployments (custom hardware and scaling control), Playground (visual model comparison), Organizations (team collaboration), Predictions (run management with webhooks), Agent Skills (AI agent builder tools with MCP server), and the Try for Free model collection. The platform competes with major cloud providers (Google, Amazon, Microsoft) and specializes in making AI deployment as accessible as importing an npm package.
Differentiator
Problem solved
Functional benefit
Products and services
- Replicate Platform Cloud-based machine learning platform that enables software developers to run, fine-tune, and deploy 50,000+ open-source and proprietary AI models (image, video, audio, LLM) via a unified API with automatic GPU scaling and pay-per-use compute billing.
- Cog
- LoRA Training Low-Rank Adaptation (LoRA) training product for Stable Diffusion models that enables fast fine-tuning in minutes with small (~5MB) output files; supports multiple concepts in a single image and preset configurations for faces, objects, and styles.
Quantifiable outcome
- Deploy an AI feature in a day and scale to millions of users without ML expertise
- +2 more outcomes
Companies that use Replicate
Customer profileNamed customers8 records
Segments4 records
Ideal customer profiles4 records
Replicate technology and API
TechnologyTechnology focussed Yes
API detail
- Has API
- Yes
- API docs
- API detail
Core technology
AI maturity
App detail
Integration2 records
AI capability10 records
Feature7 records
Replicate partnerships and signals
Strategic signalPartnerships
One partnership is on record.
- CloudflareflagshipCloudflare announced acquisition of Replicate in November 2025, completing in January 2026. Replicate continues as a distinct brand while being integrated into Cloudflare's developer platform (Workers AI, AI Gateway, Agent Cloud). The acquisition gives Cloudflare access to 50,000+ AI models and a unified API for AI model access. Replicate gains access to Cloudflare's global network infrastructure and enterprise customer base.
Scale indicators5 records
Recent moves6 records
Expansion highlights6 records
Replicate competitors and assessment
Company assessmentDirect peers
- Hugging Face: Largest open-source model hub and inference platform; directly competes with Replicate in hosting community and partner models behind a unified API, with a comparable developer-first GTM and similar open-model catalog strategy.
- Together AI: GPU cloud purpose-built for open-source model inference and fine-tuning; offers an API for running OSS LLMs and image models with pay-per-use pricing, directly overlapping Replicate's core serving and fine-tuning offering.
- Fireworks AI: Managed inference platform for open-source and proprietary models with low-latency serving and fine-tuning; competes head-to-head with Replicate for AI developer and enterprise inference workloads.
- Modal Labs: Serverless GPU platform for running ML models and arbitrary Python code; offers one-line model execution and auto-scaling comparable to Replicate, targeting the same developer persona.
- Lambda Labs: GPU cloud provider offering on-demand A100/H100 clusters and a model API; competes with Replicate for GPU-based model serving, particularly for customers that want dedicated capacity.
- RunPod: GPU cloud platform with serverless endpoints and dedicated pods for AI inference; targets the same developer audience as Replicate with per-second GPU billing and one-click model deployment.
Emerging players
- Anyscale: Ray-based compute platform for AI workloads; overlaps with Replicate in serving and scaling ML models but skews toward teams building custom inference stacks rather than consuming a hosted model catalog.
Broad incumbents
- AWS (SageMaker / Bedrock): Hyperscaler offering both a managed ML platform (SageMaker) and a multi-model inference service (Bedrock); competes with Replicate for enterprise inference spend and bundles inference into broader AWS cloud contracts.
- Google Cloud Vertex AI: Google's enterprise AI platform offering model serving, fine-tuning, and MLOps on Google's own and OSS models; explicitly named by Replicate as a competitor for enterprise AI deployment.
- Microsoft Azure AI: Hyperscaler offering Azure Machine Learning and Azure OpenAI Service; competes with Replicate for enterprise inference contracts and integrates with the broader Microsoft enterprise software footprint.
Market position
Strengths5 records
Weaknesses5 records
Competitive moat5 records
Key risks6 records
Key highlights6 records
Customer concentration
Replicate social profiles
Digital presenceReplicate compliance and trust
Trust signalCompliance1 record
Replicate financial estimates
Financial estimateRevenue estimate
Valuation estimate
Replicate leadership team
Management profileNumber of profiles
Profiles1 record
Replicate funding detail
Funding detailFunding overview
Funding rounds4 records
Investors7 records
Funding detail is available on the Subscription and Enterprise plan.Contact sales →
Replicate M&A and investment
M&A and investmentM&A
Investments
M&A and investment is available on the Subscription and Enterprise plan.Contact sales →
Frequently asked questions about Replicate
What does Replicate do?
Replicate is a cloud-based AI model deployment platform that lets developers run, fine-tune, and deploy over 50,000 open-source and proprietary machine learning models via a single API. The platform abstracts GPU infrastructure management, auto-scaling, cold boots, and versioning, allowing software developers to integrate AI capabilities into their applications with one line of code across image, video, audio, and text modalities. Supporting tooling includes Cog, an open-source Docker-based ML model packaging utility, and LoRA-based fine-tuning for image models such as SDXL and FLUX.
Is Replicate a public or private company?
Replicate is a private company. It is classified as corporate owned and is currently acquired.
When was Replicate founded?
Replicate was founded in 2023. It employs 11 to 50 people.
Where is Replicate based?
Replicate is headquartered in San Francisco, United States, in the North America region.
How does Replicate make money?
Three revenue lines are on record. Compute/Usage-Based Billing is the primary driver. The others are prepaid Credit and payment in Arrears.
Who are Replicate's main competitors?
Direct peers on record are Hugging Face, Together AI, Fireworks AI, Modal Labs, Lambda Labs and RunPod. Anyscale is listed as an emerging player. Broad incumbents are AWS (SageMaker / Bedrock), Google Cloud Vertex AI and Microsoft Azure AI.
Does Replicate have an API?
Yes. Replicate provides a cloud API that allows developers to run machine learning models without managing infrastructure. The HTTP API can be used with any programming language, with official client libraries for Python and JavaScript. The API supports predictions with polling or webhooks for long-running models, rate limits, streaming output, and file handling via replicate.delivery domain. An OpenAPI schema is available for API specification. An MCP (Model Context Protocol) server is available for agent integrations. Developer documentation is at replicate.com/docs/reference/http.
What industry is Replicate in?
Replicate's product category is AI Inference Platform. Its primary akta.pro industry code is HDAAABAF, Model Deployment, Serving & Inference Platforms, with a secondary code of HDAAACAB, Model Hosting, Serving & Inference Platforms. Its NAICS code is 5182 and its SIC code is 7372.