Home / AI Readiness Audits / Replicate (replicate.com)

Replicate (replicate.com): AI Search Readiness

Sector: AI & ML · Category: Cloud ML Inference API · Cohort Rank: #30 of 40 · Score: 53/100 (Grade C)

Replicate (replicate.com) scores 53/100 (Grade C), ranking #30 of 40 unblocked developer platforms audited on 2026-11-14 UTC (-10.1 pts vs the 63.1/100 cohort mean). From Cloudflare’s edge, https://replicate.com/ responded in 938ms TTFB with 254 KB HTML across 1 <script> tags, emits zero Schema.org application/ld+json blocks on the root HTML document, and permits all 29/29 tracked AI crawlers (0 via explicit User-agent rules, 29 via User-agent: *). At the discovery layer, replicate.com currently serves /llms.txt (HTTP 200), and returns non-200 on /llms-full.txt (HTTP 404), /pricing.md (HTTP 404), /.well-known/mcp.json (HTTP 404), /.well-known/openapi.json (HTTP 404).

Scope & limits: Technical readiness assessment from Cloudflare edge probes on 2026-11-14 UTC; independent audit not affiliated with Replicate. Technical readiness does not guarantee AI citations or sales.

1. Technical Signal Telemetry & Discovery Status

Signal / Endpoint	Observed Result	Standard Reference
Root Edge Response	938ms TTFB · 254 KB HTML · 1 `<script>` tags	Cloudflare Edge Node (NRT)
`GET /robots.txt`	HTTP 200 · 29/29 AI bots allowed	RFC 9309 crawler access policy
`GET /sitemap.xml`	HTTP 200 · Declared: https://replicate.com/sitemap.xml	Canonical URL discovery index
`GET /llms.txt`	HTTP 200	Markdown index for ChatGPT Search, Perplexity & Claude
`GET /llms-full.txt`	HTTP 404	Single-file plain-text RAG documentation corpus
`GET /pricing.md`	HTTP 404	Deterministic tier & fee schedule
`GET /.well-known/ai.txt`	HTTP 404	Machine discovery pointer manifest
`GET /ai/summary.json`	HTTP 404	Structured JSON entity & capability profile
`GET /.well-known/mcp.json`	HTTP 404	Model Context Protocol server discovery
`GET /.well-known/agent-card.json`	HTTP 404	Agent-to-Agent (A2A v0.2.5) capability card
`GET /.well-known/openapi.json`	HTTP 404	OpenAPI 3.1 function-calling contract
`Schema.org JSON-LD`	0 detected · No `application/ld+json` on `/`	Entity disambiguation & structured offers

2. 29-Crawler /robots.txt Access Breakdown

Policy Bucket	Count	Measured User-Agents
Explicitly Allowed	0/29	`None explicitly named (uses wildcard)`
*Allowed via Wildcard (``)**	29/29	`GPTBot, OAI-SearchBot, ChatGPT-User, ClaudeBot, Claude-SearchBot, Claude-User, anthropic-ai, claude-web, PerplexityBot, Perplexity-User, Googlebot, Google-Extended, Google-CloudVertexBot, Bingbot, Applebot-Extended, Applebot, DuckAssistBot, cohere-ai, Bytespider, meta-externalagent, Meta-ExternalFetcher, facebookexternalhit, Amazonbot, AI2Bot, AI2Bot-Dolma, xAI-Bot, PetalBot, YouBot, CCBot`
Disallowed (`Disallow: /`)	0/29	`None — 0 of 29 AI crawlers blocked`

3. Missing-Artifact Diffs & Remediation

/llms-full.txt (currently HTTP 404): Publish consolidated plain-text product & architecture corpus so RAG pipelines bypass 254 KB HTML / 1 script tags.
/pricing.md (currently HTTP 404): Expose deterministic plain-text USD tier table, quotas, and enterprise terms so answer engines cite verified pricing.
/.well-known/ai.txt (currently HTTP 404): Publish machine discovery pointer indexing /llms.txt, /ai/summary.json, and agent tool manifests.
/.well-known/mcp.json (currently HTTP 404): Expose Model Context Protocol discovery card so Claude Code, Cursor, and MCP registries discover official tools.
/.well-known/agent-card.json (currently HTTP 404): Add A2A v0.2.5 Agent Card declaring capability skills, authentication mode, and rate limits.
/.well-known/openapi.json (currently HTTP 404): Serve OpenAPI 3.1 JSON contract for deterministic function-calling and WebMCP execution.
/ai/summary.json (currently HTTP 404): Serve structured JSON entity profile, product category, and canonical support endpoints.
Schema.org @graph (0 JSON-LD blocks on /): Embed linked Organization, SoftwareApplication, and FAQPage JSON-LD in root HTML.
/robots.txt (wildcard-only AI policy): Declare explicit Allow: / blocks for all 29 AI search and agent user-agents to prevent CDN/WAF ambiguity.

Inspect starter /llms.txt template for replicate.com

# Replicate

> Run open-source machine learning models with a cloud API

## Core Machine Discovery (https://replicate.com)
- [Full AI Knowledge Corpus](https://replicate.com/llms-full.txt): Single-file technical architecture & product reference
- [Deterministic Pricing](https://replicate.com/pricing.md): Verified tier & enterprise fee schedule
- [Direct Answer Blocks](https://replicate.com/answers.md): 40-60 word citable BOFU answers
- [Entity Summary JSON](https://replicate.com/ai/summary.json): Structured JSON profile & WebMCP tools (verified 2026-11-14)
- [MCP Discovery Manifest](https://replicate.com/.well-known/mcp.json): Agentic discovery endpoints
- [A2A Agent Card](https://replicate.com/.well-known/agent-card.json): Agent-to-Agent v0.2.5 capability card
- [OpenAPI 3.1 Tool Spec](https://replicate.com/.well-known/openapi.json): Deterministic agent function-calling contract
- [AI Knowledge RSS Feed](https://replicate.com/feed.xml): Machine-readable update feed

One-time $49 · 26 files · Base USDC or Monero · no signup

Download free 3-file starter for replicate.com (.txt) · EdgeScreen x402 ($0.002)

23 remaining production files

/.well-known/mcp.json, agent-card.json, openapi.json, llms-full.txt, pricing.md, answers.md, ai/*.json
Schema.org JSON-LD bundle, Cloudflare Worker edge headers, comparisons, GitHub Action CI verifier.

4. Embeddable Live Score Badge (Free README / Docs Widget)

Embed Replicate’s verified AI Search Readiness badge in GitHub READMEs, engineering docs, or status pages:

View Markdown & HTML badge code

[![AI Search Readiness](https://asostudio.dev/badge/replicate.svg)](https://asostudio.dev/audits/replicate/)

<a href="https://asostudio.dev/audits/replicate/"><img src="https://asostudio.dev/badge/replicate.svg" alt="Replicate AI Search Readiness: 53/100 (C)" width="196" height="22"></a>

5. 1-Click Live Re-Scan & Custom Domain Check

Re-run our 47-check Cloudflare edge probe against replicate.com right now, or test your own domain:

6. Self-Serve 26-File ASO Pack ($49)

Generate a tailored 26-file /llms-full.txt, /pricing.md, 29-crawler robots.txt, Schema.org @graph, and 14-route Edge Router .zip for replicate.com or any domain you operate. Instant browser download via USDC on Base (Chain 8453), or Monero (XMR — verified within 24h) via non-custodial Trocador AnonPay. Inspect free sample .zip · Terms · EdgeScreen x402 ($0.002 USDC).

7. Related Audits & Head-to-Head Comparisons

Hugging Face vs Replicate (56/100 vs 53/100)
Replicate vs Together AI (53/100 vs 68/100)

Model Context Protocol (modelcontextprotocol.io) — 79/100 (Grade B+) · Agent Tooling Standard
Cursor (cursor.com) — 68/100 (Grade B) · AI Code Editor
Together AI (together.ai) — 68/100 (Grade B) · AI Inference & GPU Cloud
ElevenLabs (elevenlabs.io) — 65/100 (Grade B) · Voice & Audio AI API

Browse all 40 unblocked platform grade-cards → · Browse all 40 head-to-head comparisons →