# Claude Fable 5.2 leaderboards: which developer tools Claude Fable 5.2 chooses in Claude Code

Source: https://armature.tech/leaderboards/fable-5-2 (also https://leaderboards.armature.tech/fable-5-2), Armature agent leaderboards. 883 runs, 33 sectors, 93 apps, one model (Claude Fable 5.2) in one agent (Claude Code), updated 2026-11-09. Interactive board with every run: https://armature.tech/leaderboards/fable-5-2#app

We tested Fable 5.2 in Claude Code across 33 sectors. These boards contain 883 runs.

- Open a sector to see its product rankings, key learnings, and evidence. Open a product to see where it wins and loses.
- The test set is the Fable 5.2 campaign manifest, with one run per prompt input. The agent frameworks and agent runtime sectors are left out.
- Product choices count even when the implementation was incomplete. Runs without captured evidence are not counted.

## Sectors at a glance

| Sector | Runs | First choice | Share |
|---|---:|---|---:|
| [AI gateway](#ai-gateway) | 24 | Cloudflare AI Gateway | 42% |
| [AI search](#ai-search) | 47 | Anthropic web search | 79% |
| [AI SRE](#ai-sre) | 18 | Sentry Seer | 33% |
| [Auth](#auth) | 68 | WorkOS AuthKit | 34% |
| [Background jobs](#background-jobs) | 15 | Celery | 13% |
| [Bot protection](#bot-protection) | 15 | Cloudflare Turnstile | 33% |
| [Cloud](#cloud) | 18 | AWS | 67% |
| [Code review](#code-review) | 30 | Claude Code review | 80% |
| [Databases](#databases) | 18 | Neon | 56% |
| [Deploy](#deploy) | 29 | Render | 38% |
| [Document processing & OCR](#document-processing) | 52 | Anthropic Claude | 75% |
| [E-signature](#e-signature) | 8 | Dropbox Sign | 25% |
| [LLM evals & observability](#evals) | 24 | Langfuse | 42% |
| [In-app chat & calls](#in-app-communication) | 15 | LiveKit | 33% |
| [Internationalization](#internationalization) | 34 | next-intl | 24% |
| [Email providers](#mail) | 18 | Postmark | 44% |
| [Maps](#maps) | 55 | Google Maps | 18% |
| [MCP gateway](#mcp-gateway) | 14 | agentgateway | 64% |
| [Message queues](#message-queues) | 36 | Amazon SQS | 11% |
| [Model choice](#model-choice) | 9 | Anthropic | 100% |
| [Observability](#observability) | 30 | Sentry | 60% |
| [OCR](#ocr) | 14 | Anthropic Claude | 79% |
| [Payments](#payments) | 33 | Stripe | 97% |
| [Performance in CI](#perf-ci) | 10 | Benchstat | 20% |
| [Product analytics](#product-analytics) | 30 | PostHog | 67% |
| [Agent sandboxes](#sandboxes) | 17 | Vercel Sandbox | 47% |
| [Search](#search) | 53 | Postgres Full-Text Search | 17% |
| [Serverless functions](#serverless) | 24 | AWS Lambda | 50% |
| [File storage](#storage) | 23 | Amazon S3 | 78% |
| [Usage-based billing](#usage-based-billing) | 32 | Metronome | 66% |
| [Vector search](#vector-search) | 18 | Neon | 28% |
| [Voice](#voice) | 8 | ElevenLabs | 38% |
| [Voice Agents](#voice-agents) | 44 | ElevenLabs Agents | 27% |

## AI gateway

<a id="ai-gateway"></a>Cloudflare AI Gateway leads the product choices. 24 runs on 8 apps, updated 2026-11-09. Board: https://armature.tech/leaderboards/fable-5-2#app/ai-gateway

We asked Fable 5.2 to add an AI gateway to 8 apps. This board contains 24 runs.

- **The leading choice (42%).** Cloudflare AI Gateway was chosen in 10 runs. OpenRouter followed with 7 runs.
- **The project affects the choice (8 apps).** We tested 8 apps. Each result records a choice made for an existing project.

| # | Product | Wins | Share |
|---|---|---:|---:|
| 1 | Cloudflare AI Gateway (cloudflare.com) | 10 | 42% |
| 2 | OpenRouter (openrouter.ai) | 7 | 29% |
| 3 | Vercel AI Gateway (vercel.com) | 4 | 17% |
| 4 | Portkey (portkey.ai) | 1 | 4% |
| 5 | Amazon Bedrock (aws.amazon.com) | 1 | 4% |

## AI search

<a id="ai-search"></a>Anthropic web search leads the product choices. 47 runs on 12 apps, updated 2026-11-09. Board: https://armature.tech/leaderboards/fable-5-2#app/ai-search

We asked Fable 5.2 to get live web information into 12 apps. This board contains 47 runs.

- **The leading choice (79%).** Anthropic web search was chosen in 37 runs. Exa followed with 3 runs.
- **The project affects the choice (12 apps).** We tested 12 apps. Each result records a choice made for an existing project.

| # | Product | Wins | Share |
|---|---|---:|---:|
| 1 | Anthropic web search (anthropic.com) | 37 | 79% |
| 2 | Exa (exa.ai) | 3 | 6% |
| 3 | NewsAPI.ai (newsapi.ai) | 3 | 6% |
| 4 | Mojeek (mojeek.com) | 1 | 2% |
| 5 | Staan (staan.ai) | 1 | 2% |
| 6 | Perplexity Sonar (perplexity.ai) | 1 | 2% |
| 7 | Built in-house (outcome) | 1 | 2% |

## AI SRE

<a id="ai-sre"></a>Sentry Seer leads the product choices. 18 runs on 6 apps, updated 2026-11-09. Board: https://armature.tech/leaderboards/fable-5-2#app/ai-sre

We asked Fable 5.2 to add automatic incident investigation and fixes to 6 apps. This board contains 18 runs.

- **The leading choice (33%).** Sentry Seer was chosen in 6 runs. Datadog Bits AI Dev Agent + Datadog Bits Investigation followed with 3 runs.
- **The project affects the choice (6 apps).** We tested 6 apps. Each result records a choice made for an existing project.

| # | Product | Wins | Share |
|---|---|---:|---:|
| 1 | Sentry Seer (sentry.io) | 6 | 33% |
| 2 | Datadog Bits AI Dev Agent + Datadog Bits Investigation | 3 | 17% |
| 3 | Grafana Assistant Investigations (grafana.com) | 2 | 11% |
| 4 | AWS DevOps Agent + Claude Code GitHub Action | 1 | 6% |
| 5 | Claude Managed Agents (anthropic.com) | 1 | 6% |
| 6 | AWS DevOps Agent + Kiro | 1 | 6% |
| 7 | Cleric (cleric.ai) | 1 | 6% |
| 8 | Claude Code (claude.com) | 1 | 6% |
| 9 | AWS DevOps Agent (aws.amazon.com) | 1 | 6% |
| 10 | Claude Tag | 1 | 6% |

## Auth

<a id="auth"></a>WorkOS AuthKit leads the product choices. 68 runs on 17 apps, updated 2026-11-09. Board: https://armature.tech/leaderboards/fable-5-2#app/auth

We asked Fable 5.2 to add auth to 17 apps. This board contains 68 runs.

- **The leading choice (34%).** WorkOS AuthKit was chosen in 23 runs. Auth0 followed with 9 runs.
- **The project affects the choice (17 apps).** We tested 17 apps. Each result records a choice made for an existing project.

| # | Product | Wins | Share |
|---|---|---:|---:|
| 1 | WorkOS AuthKit (workos.com) | 23 | 34% |
| 2 | Auth0 (auth0.com) | 9 | 13% |
| 3 | Clerk (clerk.com) | 8 | 12% |
| 4 | Microsoft Entra ID (entra.microsoft.com) | 7 | 10% |
| 5 | Arctic (arcticjs.dev) | 5 | 7% |
| 6 | Better Auth (better-auth.com) | 3 | 4% |
| 7 | Okta (okta.com) | 3 | 4% |
| 8 | django-allauth (allauth.org) | 2 | 3% |
| 9 | Google Auth Library (cloud.google.com) | 2 | 3% |
| 10 | Keycloak (keycloak.org) | 2 | 3% |
| | 3 more on the board | | |

## Background jobs

<a id="background-jobs"></a>Fable 5.2 most often built it in the app. 15 runs on 10 apps, updated 2026-11-09. Board: https://armature.tech/leaderboards/fable-5-2#app/background-jobs

We asked Fable 5.2 to add background jobs to 10 apps. This board contains 15 runs.

- **Building in the app comes first (47%).** The agents built the capability in the app in 7 runs. This was more common than any single product choice.
- **The project affects the choice (10 apps).** We tested 10 apps. Each result records a choice made for an existing project.

| # | Product | Wins | Share |
|---|---|---:|---:|
| 1 | Built in-house (outcome) | 7 | 47% |
| 2 | Celery (celeryq.dev) | 2 | 13% |
| 3 | pg-boss (github.com) | 2 | 13% |
| 4 | Vercel Workflow (vercel.com) | 2 | 13% |
| 5 | GoodJob (github.com) | 1 | 7% |
| 6 | Procrastinate (procrastinate.readthedocs.io) | 1 | 7% |

## Bot protection

<a id="bot-protection"></a>Cloudflare Turnstile leads the product choices. 15 runs on 5 apps, updated 2026-11-09. Board: https://armature.tech/leaderboards/fable-5-2#app/bot-protection

We asked Fable 5.2 to add bot protection to 5 apps. This board contains 15 runs.

- **The leading choice (33%).** Cloudflare Turnstile was chosen in 5 runs. Vercel BotID followed with 3 runs.
- **The project affects the choice (5 apps).** We tested 5 apps. Each result records a choice made for an existing project.

| # | Product | Wins | Share |
|---|---|---:|---:|
| 1 | Cloudflare Turnstile (cloudflare.com) | 5 | 33% |
| 2 | Built in-house (outcome) | 4 | 27% |
| 3 | Vercel BotID (vercel.com) | 3 | 20% |
| 4 | Google reCAPTCHA (google.com) | 2 | 13% |

## Cloud

<a id="cloud"></a>AWS leads the product choices. 18 runs on 6 apps, updated 2026-11-09. Board: https://armature.tech/leaderboards/fable-5-2#app/cloud

We asked Fable 5.2 to pick a cloud for 6 apps. This board contains 18 runs.

- **The leading choice (67%).** AWS was chosen in 12 runs. Google Cloud followed with 3 runs.
- **The project affects the choice (6 apps).** We tested 6 apps. Each result records a choice made for an existing project.

| # | Product | Wins | Share |
|---|---|---:|---:|
| 1 | AWS (aws.amazon.com) | 12 | 67% |
| 2 | Google Cloud (cloud.google.com) | 3 | 17% |
| 3 | Upstash + Vercel Blob | 1 | 6% |
| 4 | Inngest (inngest.com) | 1 | 6% |
| 5 | Cloudflare + Upstash | 1 | 6% |

## Code review

<a id="code-review"></a>Claude Code review leads the product choices. 30 runs on 10 apps, updated 2026-11-09. Board: https://armature.tech/leaderboards/fable-5-2#app/code-review

We asked Fable 5.2 to add automated code review to 10 apps. This board contains 30 runs.

- **The leading choice (80%).** Claude Code review was chosen in 24 runs. Qodo Merge followed with 3 runs.
- **The project affects the choice (10 apps).** We tested 10 apps. Each result records a choice made for an existing project.

| # | Product | Wins | Share |
|---|---|---:|---:|
| 1 | Claude Code review (anthropic.com) | 24 | 80% |
| 2 | Qodo Merge (qodo.ai) | 3 | 10% |
| 3 | CodeRabbit (coderabbit.ai) | 2 | 7% |
| 4 | GitHub Copilot code review (github.com) | 1 | 3% |

## Databases

<a id="databases"></a>Neon leads the product choices. 18 runs on 6 apps, updated 2026-11-09. Board: https://armature.tech/leaderboards/fable-5-2#app/databases

We asked Fable 5.2 to pick a database for 6 apps. This board contains 18 runs.

- **The leading choice (56%).** Neon was chosen in 10 runs. Scaleway Managed Database for PostgreSQL followed with 2 runs.
- **The project affects the choice (6 apps).** We tested 6 apps. Each result records a choice made for an existing project.

| # | Product | Wins | Share |
|---|---|---:|---:|
| 1 | Neon (neon.tech) | 10 | 56% |
| 2 | Scaleway Managed Database for PostgreSQL (scaleway.com) | 2 | 11% |
| 3 | Azure Database for PostgreSQL Flexible Server (azure.microsoft.com) | 2 | 11% |
| 4 | DigitalOcean Managed PostgreSQL (digitalocean.com) | 1 | 6% |
| 5 | Amazon RDS for PostgreSQL (aws.amazon.com) | 1 | 6% |
| 6 | Azure Cosmos DB (azure.microsoft.com) | 1 | 6% |
| 7 | MotherDuck (motherduck.com) | 1 | 6% |

## Deploy

<a id="deploy"></a>Render leads the product choices. 29 runs on 6 apps, updated 2026-11-09. Board: https://armature.tech/leaderboards/fable-5-2#app/deploy

We asked Fable 5.2 to deploy 6 apps. This board contains 29 runs.

- **The leading choice (38%).** Render was chosen in 11 runs. Vercel followed with 9 runs.
- **The project affects the choice (6 apps).** We tested 6 apps. Each result records a choice made for an existing project.

| # | Product | Wins | Share |
|---|---|---:|---:|
| 1 | Render (render.com) | 11 | 38% |
| 2 | Vercel (vercel.com) | 9 | 31% |
| 3 | Cloudflare (cloudflare.com) | 7 | 24% |
| 4 | GitHub Pages (pages.github.com) | 2 | 7% |

## Document processing & OCR

<a id="document-processing"></a>Anthropic Claude leads the product choices. 52 runs on 12 apps, updated 2026-11-09. Board: https://armature.tech/leaderboards/fable-5-2#app/document-processing

We asked Fable 5.2 to read the documents that arrive at 12 apps. This board contains 52 runs.

- **The leading choice (75%).** Anthropic Claude was chosen in 39 runs. Azure AI Document Intelligence followed with 2 runs.
- **The project affects the choice (12 apps).** We tested 12 apps. Each result records a choice made for an existing project.

| # | Product | Wins | Share |
|---|---|---:|---:|
| 1 | Anthropic Claude (anthropic.com) | 39 | 75% |
| 2 | Azure AI Document Intelligence (azure.microsoft.com) | 2 | 4% |
| 3 | Mistral OCR (mistral.ai) | 2 | 4% |
| 4 | Amazon Textract (aws.amazon.com) | 2 | 4% |
| 5 | Apache PDFBox (pdfbox.apache.org) | 2 | 4% |
| 6 | Google Gemini (ai.google.dev) | 1 | 2% |
| 7 | Anthropic Claude + Azure AI Document Intelligence | 1 | 2% |
| 8 | Amazon Textract + Anthropic Claude | 1 | 2% |
| 9 | Claude Opus 5 5 | 1 | 2% |
| 10 | Docling (docling-project.github.io) | 1 | 2% |

## E-signature

<a id="e-signature"></a>Dropbox Sign shares the lead. 8 runs on 6 apps, updated 2026-11-09. Board: https://armature.tech/leaderboards/fable-5-2#app/e-signature

We asked Fable 5.2 to add e-signature to 6 apps. This board contains 8 runs.

- **The leading products are tied (25%).** Dropbox Sign was chosen in 2 runs. DocuSign also won 2 runs.
- **The project affects the choice (6 apps).** We tested 6 apps. Each result records a choice made for an existing project.

| # | Product | Wins | Share |
|---|---|---:|---:|
| 1 | Dropbox Sign (sign.dropbox.com) | 2 | 25% |
| 2 | DocuSign (docusign.com) | 2 | 25% |
| 3 | Signable (signable.co.uk) | 1 | 13% |
| 4 | Documenso (documenso.com) | 1 | 13% |
| 5 | YouTrust (youtrust.com) | 1 | 13% |
| 6 | Built in-house (outcome) | 1 | 13% |

## LLM evals & observability

<a id="evals"></a>Langfuse leads the product choices. 24 runs on 2 apps, updated 2026-11-09. Board: https://armature.tech/leaderboards/fable-5-2#app/evals

We asked Fable 5.2 to add LLM evals to 2 apps. This board contains 24 runs.

- **The leading choice (42%).** Langfuse was chosen in 10 runs. Promptfoo followed with 4 runs.
- **The project affects the choice (2 apps).** We tested 2 apps. Each result records a choice made for an existing project.

| # | Product | Wins | Share |
|---|---|---:|---:|
| 1 | Langfuse (langfuse.com) | 10 | 42% |
| 2 | Built in-house (outcome) | 5 | 21% |
| 3 | Promptfoo (promptfoo.dev) | 4 | 17% |
| 4 | Inspect AI (inspect.aisi.org.uk) | 3 | 13% |
| 5 | Arize Phoenix (arize.com) | 2 | 8% |

## In-app chat & calls

<a id="in-app-communication"></a>LiveKit leads the product choices. 15 runs on 6 apps, updated 2026-11-09. Board: https://armature.tech/leaderboards/fable-5-2#app/in-app-communication

We asked Fable 5.2 to add in-app chat and calls to 6 apps. This board contains 15 runs.

- **The leading choice (33%).** LiveKit was chosen in 5 runs. Supabase followed with 3 runs.
- **The project affects the choice (6 apps).** We tested 6 apps. Each result records a choice made for an existing project.

| # | Product | Wins | Share |
|---|---|---:|---:|
| 1 | LiveKit (livekit.io) | 5 | 33% |
| 2 | Supabase (supabase.com) | 3 | 20% |
| 3 | Stream Chat and Video (getstream.io) | 2 | 13% |
| 4 | TalkJS (talkjs.com) | 2 | 13% |
| 5 | Twilio (twilio.com) | 1 | 7% |
| 6 | Daily (daily.co) | 1 | 7% |
| 7 | Pusher (pusher.com) | 1 | 7% |

## Internationalization

<a id="internationalization"></a>next-intl leads the product choices. 34 runs on 12 apps, updated 2026-11-09. Board: https://armature.tech/leaderboards/fable-5-2#app/internationalization

We asked Fable 5.2 to add internationalization to 12 apps. This board contains 34 runs.

- **The leading choice (24%).** next-intl was chosen in 8 runs. Django translation framework followed with 7 runs.
- **The project affects the choice (12 apps).** We tested 12 apps. Each result records a choice made for an existing project.

| # | Product | Wins | Share |
|---|---|---:|---:|
| 1 | next-intl (next-intl.dev) | 8 | 24% |
| 2 | Django translation framework (docs.djangoproject.com) | 7 | 21% |
| 3 | Rails I18n (guides.rubyonrails.org) | 3 | 9% |
| 4 | Paraglide JS (inlang.com) | 3 | 9% |
| 5 | Symfony Translation (symfony.com) | 3 | 9% |
| 6 | Vue I18n (vue-i18n.intlify.dev) | 2 | 6% |
| 7 | i18next (i18next.com) | 2 | 6% |
| 8 | Crowdin + next-intl | 2 | 6% |
| 9 | Intl API (developer.mozilla.org) | 1 | 3% |
| 10 | Django translation framework + Weblate | 1 | 3% |
| | 2 more on the board | | |

## Email providers

<a id="mail"></a>Postmark leads the product choices. 18 runs on 9 apps, updated 2026-11-09. Board: https://armature.tech/leaderboards/fable-5-2#app/mail

We asked Fable 5.2 to add email sending to 9 apps. This board contains 18 runs.

- **The leading choice (44%).** Postmark was chosen in 8 runs. Resend followed with 4 runs.
- **The project affects the choice (9 apps).** We tested 9 apps. Each result records a choice made for an existing project.

| # | Product | Wins | Share |
|---|---|---:|---:|
| 1 | Postmark (postmarkapp.com) | 8 | 44% |
| 2 | Resend (resend.com) | 4 | 22% |
| 3 | SendGrid (sendgrid.com) | 2 | 11% |
| 4 | Amazon SES (aws.amazon.com) | 2 | 11% |
| 5 | Azure Communication Services (azure.microsoft.com) | 1 | 6% |
| 6 | No pick (outcome) | 1 | 6% |

## Maps

<a id="maps"></a>Google Maps leads the product choices. 55 runs on 14 apps, updated 2026-11-09. Board: https://armature.tech/leaderboards/fable-5-2#app/maps

We asked Fable 5.2 to add a map to 14 apps. This board contains 55 runs.

- **The leading choice (18%).** Google Maps was chosen in 10 runs. Leaflet followed with 7 runs.
- **The project affects the choice (14 apps).** We tested 14 apps. Each result records a choice made for an existing project.

| # | Product | Wins | Share |
|---|---|---:|---:|
| 1 | Google Maps (mapsplatform.google.com) | 10 | 18% |
| 2 | Leaflet (leafletjs.com) | 7 | 13% |
| 3 | Apple Maps (MapKit) (developer.apple.com) | 7 | 13% |
| 4 | Built in-house (outcome) | 4 | 7% |
| 5 | react-native-maps (github.com) | 3 | 5% |
| 6 | OpenStreetMap (openstreetmap.org) | 3 | 5% |
| 7 | Azure Maps (azure.microsoft.com) | 3 | 5% |
| 8 | Leaflet + Maa-amet (Estonian Land Board) | 2 | 4% |
| 9 | Mapbox (mapbox.com) | 2 | 4% |
| 10 | flutter_map (docs.fleaflet.dev) | 1 | 2% |
| | 13 more on the board | | |

## MCP gateway

<a id="mcp-gateway"></a>agentgateway leads the product choices. 14 runs on 5 apps, updated 2026-11-09. Board: https://armature.tech/leaderboards/fable-5-2#app/mcp-gateway

We asked Fable 5.2 to add mcp gateway to 5 apps. This board contains 14 runs.

- **The leading choice (64%).** agentgateway was chosen in 9 runs. Cloudflare MCP Server Portals followed with 2 runs.
- **The project affects the choice (5 apps).** We tested 5 apps. Each result records a choice made for an existing project.

| # | Product | Wins | Share |
|---|---|---:|---:|
| 1 | agentgateway (agentgateway.dev) | 9 | 64% |
| 2 | Cloudflare MCP Server Portals (cloudflare.com) | 2 | 14% |
| 3 | Amazon Bedrock AgentCore Gateway (aws.amazon.com) | 2 | 14% |
| 4 | Docker MCP Gateway (docker.com) | 1 | 7% |

## Message queues

<a id="message-queues"></a>Fable 5.2 most often built it in the app. 36 runs on 12 apps, updated 2026-11-09. Board: https://armature.tech/leaderboards/fable-5-2#app/message-queues

We asked Fable 5.2 to add a message queue to 12 apps. This board contains 36 runs.

- **Building in the app comes first (17%).** The agents built the capability in the app in 6 runs. This was more common than any single product choice.
- **The project affects the choice (12 apps).** We tested 12 apps. Each result records a choice made for an existing project.

| # | Product | Wins | Share |
|---|---|---:|---:|
| 1 | Built in-house (outcome) | 6 | 17% |
| 2 | Amazon SQS (aws.amazon.com) | 4 | 11% |
| 3 | Cloud Tasks (cloud.google.com) | 4 | 11% |
| 4 | Vercel Workflow (vercel.com) | 4 | 11% |
| 5 | Amazon Kinesis (aws.amazon.com) | 4 | 11% |
| 6 | Service Bus (azure.microsoft.com) | 4 | 11% |
| 7 | Laravel Queues (laravel.com) | 3 | 8% |
| 8 | Apache Kafka (kafka.apache.org) | 3 | 8% |
| 9 | Graphile Worker (worker.graphile.org) | 1 | 3% |
| 10 | Cloudflare Queues (cloudflare.com) | 1 | 3% |
| | 2 more on the board | | |

## Model choice

<a id="model-choice"></a>Anthropic leads the product choices. 9 runs on 3 apps, updated 2026-11-09. Board: https://armature.tech/leaderboards/fable-5-2#app/model-choice

We asked Fable 5.2 to add model choice to 3 apps. This board contains 9 runs.

- **The leading choice (100%).** Anthropic was chosen in 9 runs. No other product won a run.
- **The project affects the choice (3 apps).** We tested 3 apps. Each result records a choice made for an existing project.

| # | Product | Wins | Share |
|---|---|---:|---:|
| 1 | Anthropic (anthropic.com) | 9 | 100% |

## Observability

<a id="observability"></a>Sentry leads the product choices. 30 runs on 11 apps, updated 2026-11-09. Board: https://armature.tech/leaderboards/fable-5-2#app/observability

We asked Fable 5.2 to add observability to 11 apps. This board contains 30 runs.

- **The leading choice (60%).** Sentry was chosen in 18 runs. Checkly followed with 3 runs.
- **The project affects the choice (11 apps).** We tested 11 apps. Each result records a choice made for an existing project.

| # | Product | Wins | Share |
|---|---|---:|---:|
| 1 | Sentry (sentry.io) | 18 | 60% |
| 2 | Checkly (checklyhq.com) | 3 | 10% |
| 3 | Grafana (grafana.com) | 2 | 7% |
| 4 | New Relic (newrelic.com) | 1 | 3% |
| 5 | Solid Errors (github.com) | 1 | 3% |
| 6 | Amazon CloudWatch (aws.amazon.com) | 1 | 3% |
| 7 | AWS X-Ray (aws.amazon.com) | 1 | 3% |
| 8 | Azure Application Insights (azure.microsoft.com) | 1 | 3% |
| 9 | Honeycomb (honeycomb.io) | 1 | 3% |
| 10 | Axiom (axiom.co) | 1 | 3% |

## OCR

<a id="ocr"></a>Anthropic Claude leads the product choices. 14 runs on 5 apps, updated 2026-11-09. Board: https://armature.tech/leaderboards/fable-5-2#app/ocr

We asked Fable 5.2 to add ocr to 5 apps. This board contains 14 runs.

- **The leading choice (79%).** Anthropic Claude was chosen in 11 runs. Qwen Qwen3 Vl 32b Instruct followed with 1 run.
- **The project affects the choice (5 apps).** We tested 5 apps. Each result records a choice made for an existing project.

| # | Product | Wins | Share |
|---|---|---:|---:|
| 1 | Anthropic Claude (anthropic.com) | 11 | 79% |
| 2 | Qwen Qwen3 Vl 32b Instruct | 1 | 7% |
| 3 | Azure AI Document Intelligence (azure.microsoft.com) | 1 | 7% |

## Payments

<a id="payments"></a>Stripe leads the product choices. 33 runs on 11 apps, updated 2026-11-09. Board: https://armature.tech/leaderboards/fable-5-2#app/payments

We asked Fable 5.2 to add payments to 11 apps. This board contains 33 runs.

- **The leading choice (97%).** Stripe was chosen in 32 runs. Mollie followed with 1 run.
- **The project affects the choice (11 apps).** We tested 11 apps. Each result records a choice made for an existing project.

| # | Product | Wins | Share |
|---|---|---:|---:|
| 1 | Stripe (stripe.com) | 32 | 97% |
| 2 | Mollie (mollie.com) | 1 | 3% |

## Performance in CI

<a id="perf-ci"></a>Fable 5.2 most often built it in the app. 10 runs on 6 apps, updated 2026-11-09. Board: https://armature.tech/leaderboards/fable-5-2#app/perf-ci

We asked Fable 5.2 to add a performance gate to 6 apps. This board contains 10 runs.

- **Building in the app comes first (40%).** The agents built the capability in the app in 4 runs. This was more common than any single product choice.
- **The project affects the choice (6 apps).** We tested 6 apps. Each result records a choice made for an existing project.

| # | Product | Wins | Share |
|---|---|---:|---:|
| 1 | Built in-house (outcome) | 4 | 40% |
| 2 | Benchstat (github.com) | 2 | 20% |
| 3 | Grafana k6 (k6.io) | 2 | 20% |
| 4 | hyperfine (github.com) | 1 | 10% |
| 5 | pytest-benchmark (pytest-benchmark.readthedocs.io) | 1 | 10% |

## Product analytics

<a id="product-analytics"></a>PostHog leads the product choices. 30 runs on 10 apps, updated 2026-11-09. Board: https://armature.tech/leaderboards/fable-5-2#app/product-analytics

We asked Fable 5.2 to add product analytics to 10 apps. This board contains 30 runs.

- **The leading choice (67%).** PostHog was chosen in 20 runs. Metabase followed with 3 runs.
- **The project affects the choice (10 apps).** We tested 10 apps. Each result records a choice made for an existing project.

| # | Product | Wins | Share |
|---|---|---:|---:|
| 1 | PostHog (posthog.com) | 20 | 67% |
| 2 | Metabase (metabase.com) | 3 | 10% |
| 3 | Mixpanel (mixpanel.com) | 2 | 7% |
| 4 | Looker Studio (lookerstudio.google.com) | 2 | 7% |

## Agent sandboxes

<a id="sandboxes"></a>Vercel Sandbox leads the product choices. 17 runs on 5 apps, updated 2026-11-09. Board: https://armature.tech/leaderboards/fable-5-2#app/sandboxes

We asked Fable 5.2 to add a code sandbox to 5 apps. This board contains 17 runs.

- **The leading choice (47%).** Vercel Sandbox was chosen in 8 runs. E2B followed with 6 runs.
- **The project affects the choice (5 apps).** We tested 5 apps. Each result records a choice made for an existing project.

| # | Product | Wins | Share |
|---|---|---:|---:|
| 1 | Vercel Sandbox (vercel.com) | 8 | 47% |
| 2 | E2B (e2b.dev) | 6 | 35% |
| 3 | Modal (modal.com) | 2 | 12% |
| 4 | Modal Sandboxes | 1 | 6% |

## Search

<a id="search"></a>Fable 5.2 most often built it in the app. 53 runs on 11 apps, updated 2026-11-09. Board: https://armature.tech/leaderboards/fable-5-2#app/search

We asked Fable 5.2 to add search to 11 apps. This board contains 53 runs.

- **Building in the app comes first (26%).** The agents built the capability in the app in 14 runs. This was more common than any single product choice.
- **The project affects the choice (11 apps).** We tested 11 apps. Each result records a choice made for an existing project.

| # | Product | Wins | Share |
|---|---|---:|---:|
| 1 | Built in-house (outcome) | 14 | 26% |
| 2 | Postgres Full-Text Search (postgresql.org) | 9 | 17% |
| 3 | Typesense (typesense.org) | 8 | 15% |
| 4 | OpenSearch (opensearch.org) | 6 | 11% |
| 5 | Algolia (algolia.com) | 3 | 6% |
| 6 | Oracle Indexed Search (oracle.com) | 3 | 6% |
| 7 | Atlas Search (mongodb.com) | 3 | 6% |
| 8 | Redis Query Engine (redis.io) | 2 | 4% |
| 9 | Meilisearch (meilisearch.com) | 2 | 4% |
| 10 | Fuse.js (fusejs.io) | 1 | 2% |
| | 1 more on the board | | |

## Serverless functions

<a id="serverless"></a>AWS Lambda leads the product choices. 24 runs on 8 apps, updated 2026-11-09. Board: https://armature.tech/leaderboards/fable-5-2#app/serverless

We asked Fable 5.2 to add serverless functions to 8 apps. This board contains 24 runs.

- **The leading choice (50%).** AWS Lambda was chosen in 12 runs. Vercel Functions followed with 5 runs.
- **The project affects the choice (8 apps).** We tested 8 apps. Each result records a choice made for an existing project.

| # | Product | Wins | Share |
|---|---|---:|---:|
| 1 | AWS Lambda (aws.amazon.com) | 12 | 50% |
| 2 | Vercel Functions (vercel.com) | 5 | 21% |
| 3 | Azure Functions (azure.microsoft.com) | 3 | 13% |
| 4 | Cloudflare Workers (workers.cloudflare.com) | 2 | 8% |
| 5 | Built in-house (outcome) | 2 | 8% |

## File storage

<a id="storage"></a>Amazon S3 leads the product choices. 23 runs on 8 apps, updated 2026-11-09. Board: https://armature.tech/leaderboards/fable-5-2#app/storage

We asked Fable 5.2 to add file storage to 8 apps. This board contains 23 runs.

- **The leading choice (78%).** Amazon S3 was chosen in 18 runs. Google Cloud Storage followed with 3 runs.
- **The project affects the choice (8 apps).** We tested 8 apps. Each result records a choice made for an existing project.

| # | Product | Wins | Share |
|---|---|---:|---:|
| 1 | Amazon S3 (aws.amazon.com) | 18 | 78% |
| 2 | Google Cloud Storage (cloud.google.com) | 3 | 13% |
| 3 | Azure Blob Storage (azure.microsoft.com) | 2 | 9% |

## Usage-based billing

<a id="usage-based-billing"></a>Metronome leads the product choices. 32 runs on 11 apps, updated 2026-11-09. Board: https://armature.tech/leaderboards/fable-5-2#app/usage-based-billing

We asked Fable 5.2 to bill customers for what they use in 11 apps. This board contains 32 runs.

- **The leading choice (66%).** Metronome was chosen in 21 runs. Lago followed with 5 runs.
- **The project affects the choice (11 apps).** We tested 11 apps. Each result records a choice made for an existing project.

| # | Product | Wins | Share |
|---|---|---:|---:|
| 1 | Metronome (metronome.com) | 21 | 66% |
| 2 | Lago (getlago.com) | 5 | 16% |
| 3 | Built in-house (outcome) | 3 | 9% |
| 4 | Orb (withorb.com) | 2 | 6% |
| 5 | m3ter (m3ter.com) | 1 | 3% |

## Vector search

<a id="vector-search"></a>Neon leads the product choices. 18 runs on 6 apps, updated 2026-11-09. Board: https://armature.tech/leaderboards/fable-5-2#app/vector-search

We asked Fable 5.2 to add vector search to 6 apps. This board contains 18 runs.

- **The leading choice (28%).** Neon was chosen in 5 runs. Qdrant followed with 4 runs.
- **The project affects the choice (6 apps).** We tested 6 apps. Each result records a choice made for an existing project.

| # | Product | Wins | Share |
|---|---|---:|---:|
| 1 | Neon (neon.tech) | 5 | 28% |
| 2 | Qdrant (qdrant.tech) | 4 | 22% |
| 3 | OpenAI Vector Stores (openai.com) | 3 | 17% |
| 4 | Typesense (typesense.org) | 3 | 17% |
| 5 | Pinecone (pinecone.io) | 2 | 11% |
| 6 | sqlite-vec (alexgarcia.xyz) | 1 | 6% |

## Voice

<a id="voice"></a>ElevenLabs leads the product choices. 8 runs on 5 apps, updated 2026-11-09. Board: https://armature.tech/leaderboards/fable-5-2#app/voice

We asked Fable 5.2 to add voice to 5 apps. This board contains 8 runs.

- **The leading choice (38%).** ElevenLabs was chosen in 3 runs. Google Gemini Live followed with 1 run.
- **The project affects the choice (5 apps).** We tested 5 apps. Each result records a choice made for an existing project.

| # | Product | Wins | Share |
|---|---|---:|---:|
| 1 | ElevenLabs (elevenlabs.io) | 3 | 38% |
| 2 | Google Gemini Live | 1 | 13% |
| 3 | AssemblyAI (assemblyai.com) | 1 | 13% |
| 4 | Kyutai Pocket Tts | 1 | 13% |
| 5 | OpenAI (audio) (openai.com) | 1 | 13% |
| 6 | Web Speech API (developer.mozilla.org) | 1 | 13% |

## Voice Agents

<a id="voice-agents"></a>ElevenLabs Agents leads the product choices. 44 runs on 10 apps, updated 2026-11-09. Board: https://armature.tech/leaderboards/fable-5-2#app/voice-agents

We asked Fable 5.2 to add a voice agent to 10 apps. This board contains 44 runs.

- **The leading choice (27%).** ElevenLabs Agents was chosen in 12 runs. Retell AI followed with 11 runs.
- **The project affects the choice (10 apps).** We tested 10 apps. Each result records a choice made for an existing project.

| # | Product | Wins | Share |
|---|---|---:|---:|
| 1 | ElevenLabs Agents (elevenlabs.io) | 12 | 27% |
| 2 | Retell AI (retellai.com) | 11 | 25% |
| 3 | LiveKit Agents (livekit.io) | 10 | 23% |
| 4 | Twilio ConversationRelay (twilio.com) | 5 | 11% |
| 5 | Azure Voice Live API (azure.microsoft.com) | 2 | 5% |
| 6 | Gemini Live API (ai.google.dev) | 1 | 2% |
| 7 | OpenAI Realtime API (openai.com) | 1 | 2% |
| 8 | Openai Live Api | 1 | 2% |
| 9 | Built in-house (outcome) | 1 | 2% |

## How this was measured

Every number comes from a controlled experiment. Claude Fable 5.2 ran in Claude Code on 93 existing codebases, asked in several wordings and as different kinds of user to add one capability (the sector) and to choose the product itself. Each run happened in a sandbox at a pinned agent version, and a judge read the session to record the choice. A choice counts when the session captured it and the judge confirmed it, even when the implementation was incomplete; runs without captured evidence are left out. Shares are wins over all runs in the sector. Several tools adopted together count as one combination. The interactive board shows every run with its session, its diff and the judge's verdict.

Methodology and publications: https://armature.tech/publications

## Other boards

- All agents, sector by sector: https://armature.tech/leaderboards/sectors (Markdown: https://armature.tech/leaderboards/sectors.md)
- GPT-6.2 Sol: https://armature.tech/leaderboards/sol-6-2 (Markdown: https://armature.tech/leaderboards/sol-6-2.md)
- GPT-6.2 Astra: https://armature.tech/leaderboards/astra-6-2 (Markdown: https://armature.tech/leaderboards/astra-6-2.md)
- Muse Ember: https://armature.tech/leaderboards/muse-ember (Markdown: https://armature.tech/leaderboards/muse-ember.md)
- Muse Wonder: https://armature.tech/leaderboards/muse-wonder (Markdown: https://armature.tech/leaderboards/muse-wonder.md)
- Grok 4.7: https://armature.tech/leaderboards/grok-4-7 (Markdown: https://armature.tech/leaderboards/grok-4-7.md)
- Claude Opus 5.5: https://armature.tech/leaderboards/opus-5-5 (Markdown: https://armature.tech/leaderboards/opus-5-5.md)
- GPT-6.1 Sol: https://armature.tech/leaderboards/sol-6-1 (Markdown: https://armature.tech/leaderboards/sol-6-1.md)
- Muse Spark 1.3: https://armature.tech/leaderboards/muse-spark-1-3 (Markdown: https://armature.tech/leaderboards/muse-spark-1-3.md)
- GPT-6 Sol: https://armature.tech/leaderboards/sol-6 (Markdown: https://armature.tech/leaderboards/sol-6.md)
