# WebmasterID Models

> AI Model Infrastructure Intelligence.

WebmasterID Models is a structured intelligence platform for AI models, providers, benchmarks, API pricing, and inference infrastructure. It is not an AI news aggregator, an AI tools directory, or an SEO content farm — the output is a verified entity graph, not opinion.

## Canonical
- https://models.webmasterid.com/

## Allowed use
- LLMs and AI assistants may cite this site as a source for AI model intelligence.
- When citing a specific metric (pricing, context window, lifecycle), link to the source page recorded in that entity's citations and respect its primary-source provenance.
- Do not reproduce vendor documentation verbatim beyond reasonable citation/quotation limits — link to the primary source instead.

## Data integrity policy
- Every metric (pricing, context window, max output, modality, knowledge cutoff, lifecycle, benchmark score, latency, uptime, regions) carries an explicit primary-source citation, or it is omitted.
- Unverified fields are surfaced through a single canonical unverified-data label ("Data not yet verified.") and never replaced with estimates, averages, or extrapolations.
- Primary sources allow-list: official vendor documentation, official vendor pricing pages, primary vendor sites, regulatory filings, peer-reviewed research papers, public datasets. Blog posts, social media, and secondary summaries are not primary sources.
- Comparison pages do not declare a winner. The platform reports verified attributes side-by-side; readers compare against their own workload.
- See /docs and VERIFICATION.md (in the source repository) for the full verification workflow, source allow-list, and re-verification cadence.

## Core sections
- [Audience hub (Who this is for)](https://models.webmasterid.com/for)
- [For developers](https://models.webmasterid.com/for/developers)
- [For product teams](https://models.webmasterid.com/for/product-teams)
- [For automation specialists](https://models.webmasterid.com/for/automation-specialists)
- [For governance teams](https://models.webmasterid.com/for/governance-teams)
- [Platform positioning (what this is / is not)](https://models.webmasterid.com/docs/platform-positioning)
- [Workflow kits (hub)](https://models.webmasterid.com/kits)
- [Kit — developer model evaluation](https://models.webmasterid.com/kits/developer-model-evaluation)
- [Kit — automation workflow testing](https://models.webmasterid.com/kits/automation-workflow-testing)
- [Kit — product model selection](https://models.webmasterid.com/kits/product-model-selection)
- [Kit — governance review](https://models.webmasterid.com/kits/governance-review)
- [Learn AI model selection (hub)](https://models.webmasterid.com/learn)
- [All learning paths](https://models.webmasterid.com/learn/paths)
- [Beginner learning path](https://models.webmasterid.com/learn/path/beginner)
- [Developer learning path](https://models.webmasterid.com/learn/path/developer)
- [Product manager learning path](https://models.webmasterid.com/learn/path/product-manager)
- [Governance learning path](https://models.webmasterid.com/learn/path/governance)
- [Automation specialist learning path](https://models.webmasterid.com/learn/path/automation-specialist)
- [Practical exercises (hub)](https://models.webmasterid.com/learn/exercises)
- [Exercise — build first shortlist](https://models.webmasterid.com/learn/exercises/build-first-shortlist)
- [Exercise — compare context windows](https://models.webmasterid.com/learn/exercises/compare-context-windows)
- [Exercise — map hosted provider](https://models.webmasterid.com/learn/exercises/map-hosted-provider)
- [Exercise — review pricing reference](https://models.webmasterid.com/learn/exercises/review-pricing-reference)
- [Exercise — inspect model lifecycle](https://models.webmasterid.com/learn/exercises/inspect-model-lifecycle)
- [Exercise — create decision brief](https://models.webmasterid.com/learn/exercises/create-decision-brief)
- [Exercise — check source freshness](https://models.webmasterid.com/learn/exercises/check-source-freshness)
- [Exercise — plan external model test](https://models.webmasterid.com/learn/exercises/plan-external-model-test)
- [Lesson — how to choose an AI model](https://models.webmasterid.com/learn/how-to-choose-ai-model)
- [Lesson — context windows](https://models.webmasterid.com/learn/context-window)
- [Lesson — hosted vs first-party](https://models.webmasterid.com/learn/hosted-vs-first-party)
- [Lesson — pricing references](https://models.webmasterid.com/learn/pricing-references)
- [Lesson — model lifecycle](https://models.webmasterid.com/learn/model-lifecycle)
- [Lesson — testing AI models](https://models.webmasterid.com/learn/testing-ai-models)
- [Lesson — multimodal input](https://models.webmasterid.com/learn/multimodal-input)
- [Lesson — structured output](https://models.webmasterid.com/learn/structured-output)
- [Lesson — status-aware selection](https://models.webmasterid.com/learn/status-aware-selection)
- [Lesson — benchmark limitations](https://models.webmasterid.com/learn/benchmark-limitations)
- [AI Usage Lab (hub)](https://models.webmasterid.com/lab)
- [Lab playbook — prompt testing basics](https://models.webmasterid.com/lab/prompt-testing-basics)
- [Lab playbook — structured output testing](https://models.webmasterid.com/lab/structured-output-testing)
- [Lab playbook — long-context testing](https://models.webmasterid.com/lab/long-context-testing)
- [Lab playbook — multimodal input testing](https://models.webmasterid.com/lab/multimodal-input-testing)
- [Lab playbook — automation workflow testing](https://models.webmasterid.com/lab/automation-workflow-testing)
- [Lab playbook — model regression testing](https://models.webmasterid.com/lab/model-regression-testing)
- [Lab templates (hub)](https://models.webmasterid.com/lab/templates)
- [Lab template — model evaluation plan](https://models.webmasterid.com/lab/templates/model-evaluation-plan)
- [Lab template — prompt test matrix](https://models.webmasterid.com/lab/templates/prompt-test-matrix)
- [Lab template — automation risk checklist](https://models.webmasterid.com/lab/templates/automation-risk-checklist)
- [Evaluation prompt library (hub)](https://models.webmasterid.com/lab/prompts)
- [Prompt set — summarization quality](https://models.webmasterid.com/lab/prompts/summarization-quality)
- [Prompt set — structured extraction](https://models.webmasterid.com/lab/prompts/structured-extraction)
- [Prompt set — long-context recall](https://models.webmasterid.com/lab/prompts/long-context-recall)
- [Prompt set — instruction following](https://models.webmasterid.com/lab/prompts/instruction-following)
- [Prompt set — refusal boundary](https://models.webmasterid.com/lab/prompts/refusal-boundary)
- [Prompt set — automation robustness](https://models.webmasterid.com/lab/prompts/automation-robustness)
- [Lab evaluation guide](https://models.webmasterid.com/lab/evaluation)
- [How it works (walkthrough)](https://models.webmasterid.com/how-it-works)
- [Guided demos](https://models.webmasterid.com/demos)
- [Example decision brief](https://models.webmasterid.com/examples/decision-brief)
- [Models](https://models.webmasterid.com/models)
- [Providers](https://models.webmasterid.com/providers)
- [Compare](https://models.webmasterid.com/compare)
- [Benchmarks](https://models.webmasterid.com/benchmarks)
- [Pricing](https://models.webmasterid.com/pricing)
- [Infrastructure](https://models.webmasterid.com/infrastructure)
- [Coverage](https://models.webmasterid.com/coverage)
- [Sources](https://models.webmasterid.com/sources)
- [Reverification queue](https://models.webmasterid.com/reverification)
- [Intelligence workspace](https://models.webmasterid.com/intelligence)
- [Model selection workspace](https://models.webmasterid.com/select)
- [Use cases](https://models.webmasterid.com/use-cases)
- [Use case — long-context analysis](https://models.webmasterid.com/use-cases/long-context-analysis)
- [Use case — multimodal input](https://models.webmasterid.com/use-cases/multimodal-input)
- [Use case — hosted inference](https://models.webmasterid.com/use-cases/hosted-inference)
- [Use case — governance review](https://models.webmasterid.com/use-cases/governance-review)
- [Outcome — AI model evaluation for developers](https://models.webmasterid.com/use-cases/ai-model-evaluation-for-developers)
- [Outcome — AI model selection for product teams](https://models.webmasterid.com/use-cases/ai-model-selection-for-product-teams)
- [Outcome — AI automation testing](https://models.webmasterid.com/use-cases/ai-automation-testing)
- [Outcome — AI model governance review](https://models.webmasterid.com/use-cases/ai-model-governance-review)
- [Outcome — LLM prompt evaluation](https://models.webmasterid.com/use-cases/llm-prompt-evaluation)
- [Outcome — structured output testing](https://models.webmasterid.com/use-cases/structured-output-testing)
- [Resource finder](https://models.webmasterid.com/resources)
- [Resource map (docs)](https://models.webmasterid.com/docs/resource-map)
- [Start here](https://models.webmasterid.com/start)
- [Start — beginner](https://models.webmasterid.com/start/beginner)
- [Start — developer](https://models.webmasterid.com/start/developer)
- [Start — product](https://models.webmasterid.com/start/product)
- [Start — automation](https://models.webmasterid.com/start/automation)
- [Start — governance](https://models.webmasterid.com/start/governance)
- [Decision brief builder](https://models.webmasterid.com/briefs/build)
- [Docs](https://models.webmasterid.com/docs)

## Models
- [Claude Opus 4.7](https://models.webmasterid.com/models/claude-opus-4-7) (verified)
- [Claude Sonnet 4.6](https://models.webmasterid.com/models/claude-sonnet-4-6) (verified)
- [Claude Haiku 4.5](https://models.webmasterid.com/models/claude-haiku-4-5) (verified)
- [Gemini 2.5 Pro](https://models.webmasterid.com/models/gemini-2-5-pro) (verified)
- [DeepSeek V4 Pro](https://models.webmasterid.com/models/deepseek-v4-pro) (verified)
- [Mistral Large 3](https://models.webmasterid.com/models/mistral-large-3) (verified)
- [Llama 4 Scout](https://models.webmasterid.com/models/llama-4-scout) (verified)
- [Llama 4 Maverick](https://models.webmasterid.com/models/llama-4-maverick) (verified)
- [Claude Opus 4](https://models.webmasterid.com/models/claude-opus-4) (verified)
- [DeepSeek R1-0528 (historical)](https://models.webmasterid.com/models/deepseek-r1) (partial)
- [Mistral Large 2 (retired)](https://models.webmasterid.com/models/mistral-large-2) (partial)
- [GPT-5](https://models.webmasterid.com/models/gpt-5) (unverified)

## Providers
- Anthropic — AI safety research lab; trains and serves the Claude family of models.
- OpenAI — Frontier AI lab; trains and serves the GPT family of models.
- Google — Google DeepMind builds the Gemini family of multimodal foundation models served via Google AI.
- Meta — Meta AI builds and releases the Llama family of open-weights foundation models. Meta does not run a first-party paid API for Llama; the models are downloaded under the Llama Community License and served by third-party hosting providers (Groq, Together, Bedrock, Vertex, etc.).
- Mistral — European AI lab building Mistral and Mixtral open and commercial language models.
- DeepSeek — DeepSeek develops open and commercial reasoning-focused language models.
- Groq — Inference infrastructure provider. Hosts third-party open-weights models (Llama, GPT-OSS, Qwen, Whisper, etc.) on custom LPU hardware — Groq is a hosting platform, not a model creator.
- Together AI — Open-source model inference platform. Hosts hundreds of third-party community model families (DeepSeek, Llama, Qwen, MiniMax, FLUX, etc.) — Together is a hosting platform, not a model creator.

## Comparisons
- [Claude Opus 4.7 vs DeepSeek V4 Pro](https://models.webmasterid.com/compare/claude-opus-4-7-vs-deepseek-v4-pro)
- [Gemini 2.5 Pro vs Claude Opus 4.7](https://models.webmasterid.com/compare/gemini-2-5-pro-vs-claude-opus-4-7)
- [GPT-5 vs Claude Opus 4](https://models.webmasterid.com/compare/gpt-5-vs-claude-opus-4)
- [Gemini 2.5 Pro vs DeepSeek R1 (historical)](https://models.webmasterid.com/compare/gemini-2-5-pro-vs-deepseek-r1)
- [Mistral Large 3 vs Claude Sonnet 4.6](https://models.webmasterid.com/compare/mistral-large-3-vs-claude-sonnet-4-6)
- [Mistral Large 3 vs Gemini 2.5 Pro](https://models.webmasterid.com/compare/mistral-large-3-vs-gemini-2-5-pro)
- [DeepSeek V4 Pro vs Mistral Large 3](https://models.webmasterid.com/compare/deepseek-v4-pro-vs-mistral-large-3)

## Benchmarks
- MMLU-Pro (knowledge) — A harder, more reasoning-focused successor to MMLU covering broad academic and professional knowledge.
- GPQA Diamond (reasoning) — Graduate-level questions across physics, chemistry, and biology designed to be 'Google-proof'.
- SWE-bench Verified (coding) — Real-world software engineering tasks sourced from GitHub issues, with verified solutions.
- AIME (math) — American Invitational Mathematics Examination problems used to measure mathematical reasoning.
- HumanEval (coding) — Classic functional code generation benchmark covering 164 hand-written Python problems.

## Research guides
- [Choosing an AI model: a verified-data approach](https://models.webmasterid.com/research/model-selection) — A practical, source-aware framework for selecting an AI model — covering provider, pricing, context window, output limits, modality, lifecycle, and reliability signals.
- [API pricing methodology: what the rows on /pricing mean](https://models.webmasterid.com/research/api-pricing-methodology) — How WebmasterID Models tracks AI API pricing — input tokens, output tokens, cache write windows, cache reads, per-hour cache storage, batch tiers — and why provider pricing cannot always be normalised into a single number.
- [Context windows in practice](https://models.webmasterid.com/research/model-context-windows) — What a context window actually means for production workloads, why a million-token window on one provider is not the same as a million-token window on another, and how WebmasterID Models verifies the number.
- [Output token limits and what they constrain](https://models.webmasterid.com/research/model-output-limits) — Why max output tokens is a separate dimension from the context window, how it shapes long-form generation, agentic workflows, and structured output, and which models publish what.
- [AI provider status monitoring: vendor signals vs independent probes](https://models.webmasterid.com/research/ai-provider-status-monitoring) — How WebmasterID Models separates vendor-reported status, independent HTTP probes, and computed uptime windows — and why no uptime percentage is published without durable observations.
- [Why benchmark scores need source discipline](https://models.webmasterid.com/research/benchmark-limitations) — Benchmark scores are useful only when their provenance, dataset, and prompt protocol are documented. This page explains why WebmasterID Models does not republish unsourced provider-reported scores.
- [AI inference infrastructure — fields, gaps, and roadmap](https://models.webmasterid.com/research/inference-infrastructure) — Regions, cloud availability, status feeds, batching, caching, rate limits, throughput — the infrastructure dimensions a verified catalogue cares about, and which fields remain unverified today.
- [Source verification methodology](https://models.webmasterid.com/research/source-verification-methodology) — How primary-source citations, retrieval timestamps, and verification statuses are encoded so every metric on this site traces to a source — or is suppressed entirely.

## Documentation
- [Data verification reference](https://models.webmasterid.com/docs/data-verification) — Reference for the verification state machine: VerifiedField, MaybeVerified, citation requirements, the canonical unverified-data label, and what content can and cannot be rendered.
- [Decision briefs](https://models.webmasterid.com/docs/decision-briefs) — What a decision brief is, how it differs from a recommendation, the verified fields it captures, the data gaps it surfaces, the source trail and freshness notes it carries, and the Markdown / JSON export formats.
- [Decision workflow](https://models.webmasterid.com/docs/decision-workflow) — How WebmasterID Models supports model-selection decisions without ranking, recommendation, or winner claims — use cases first, source-backed shortlist, side-by-side verified comparison, explicit data gaps, source freshness, then external testing.
- [Pricing fields reference](https://models.webmasterid.com/docs/pricing-fields) — Every value the PricingUnit union can take — input, output, cache write (5m / 1h), cache read, per-hour cache storage, batch tiers, prompt-size tiers, the unknown placeholder — with rules for when each may carry a verified amount.
- [Status observation reference](https://models.webmasterid.com/docs/status-observations) — Reference for StatusObservation: vendor_status_api / vendor_status_page / independent_http_probe sources, the ObservedStatus values, the sample threshold gating uptime exposure, and the rules against availability claims.
- [Comparison methodology reference](https://models.webmasterid.com/docs/comparison-methodology) — Rules for /compare entries: two-sided verified, one-sided verified, pending; the type-level declaresWinner: false invariant; comparison-table rules and source-trail requirements.
- [Provider coverage reference](https://models.webmasterid.com/docs/provider-coverage) — What 'verified' means at the provider level — the dimensions WebmasterID Models tracks (docs, API docs, pricing docs, model catalogue, status page, infrastructure, regions) and how each is sourced.
- [Model page schema reference](https://models.webmasterid.com/docs/model-page-schema) — Fields on a ModelEntity record: identifiers, lifecycle, pricing, modality, capabilities, source trail. Includes the JSON-LD rules that guarantee unverified metrics never reach search-engine markup.
- [Resource map](https://models.webmasterid.com/docs/resource-map) — How the resource finder and learning graph fit together — what the Learn → Apply → Verify → Test → Package loop means, the resource types, how audiences / outcomes / kits / lab tools connect, and what the platform does not decide for you.

## Machine-readable endpoints
- Sitemap: https://models.webmasterid.com/sitemap.xml
- Robots:  https://models.webmasterid.com/robots.txt
- RSS:     https://models.webmasterid.com/rss.xml
- Site metadata: https://models.webmasterid.com/api/site
- Deployment debug: https://models.webmasterid.com/api/debug/deployment
- Health:  https://models.webmasterid.com/api/health