Workspace · step 5 of 5

Decision Brief Builder

Export a paste-ready evidence pack for 2–4 selected models — verified evidence rows, explicit data gaps, the source trail, freshness notes, hosted availability, and a checklist of external tests still to run. Markdown by default; JSON via the export endpoint. Briefs are evidence, not recommendations. Generated query URLs are noindex; only the base /briefs/build page is indexable. New here? Start at /how-it-works.

Decision briefs are evidence packs, not recommendations.

A brief collects verified fields, data gaps, source trails, freshness notes, and hosted availability for 2–4 selected models. WebmasterID Models does not declare a winner, does not rank by price, and does not generate conclusions that imply endorsement. Unknown values stay explicitly unverified.

Not sure what a brief looks like? View an example decision brief → (pre-built from the long-context-analysis demo).

Plan the tests that produce the evidence in this brief at /lab — six testing playbooks + paste-ready Markdown templates. Use /lab/prompts to collect external observations before finalising this brief.

Confused by what to put in a brief? /learn/testing-ai-models explains how the brief pairs with your own tests. Or follow a sequenced path: Product manager · Governance.

ResetCap: 4 models per brief.

Selected models

Verified evidence rows

Data gaps

Source citations

Hosted availability

Models

Selected models (4)

In query order. Lifecycle reflects the verified value or 'unverified' when null.

Gemini 2.5 Pro
Google · lifecycle: active
https://models.webmasterid.com/models/gemini-2-5-pro
Claude Haiku 4.5
Anthropic · lifecycle: active
https://models.webmasterid.com/models/claude-haiku-4-5
Claude Opus 4.7
Anthropic · lifecycle: active
https://models.webmasterid.com/models/claude-opus-4-7
Claude Sonnet 4.6
Anthropic · lifecycle: active
https://models.webmasterid.com/models/claude-sonnet-4-6

Evidence

Verified evidence (32)

Per-model verified field values. Source IDs map to the source trail below.

Model	Field	Value	Sources
gemini-2-5-pro	identity	Google (gemini-2.5-pro)	`ai.google.dev/gemini-api`
gemini-2-5-pro	lifecycle	active	`ai.google.dev/gemini-api`
gemini-2-5-pro	context	1,048,576 tokens	`ai.google.dev/gemini-api`
gemini-2-5-pro	output	65,536 tokens max output	`ai.google.dev/gemini-api`
gemini-2-5-pro	modality	text-in, image-in, audio-in, video-in, text-out	`ai.google.dev/gemini-api`
gemini-2-5-pro	pricing	1M input tokens: $1.25 · 1M input tokens (>200k context): $2.5 · 1M output tokens: $10 · 1M output tokens (>200k context): $15 · 1M cache write tokens (5m): $0.125 · 1M cache write tokens (>200k context): $0.25 · 1M cache storage / hour: $4.5 · 1M batch input tokens: $0.625 · 1M batch input tokens (>200k context): $1.25 · 1M batch output tokens: $5 · 1M batch output tokens (>200k context): $7.5 (reference, not live quote)	`ai.google.dev/pricing`
gemini-2-5-pro	sources	3 primary-source citations	`ai.google.dev/gemini-apiai.google.dev/pricingai.google.dev/gemini-api`
gemini-2-5-pro	freshness	Fresh (last checked 2026-05-20)	—
claude-haiku-4-5	identity	Anthropic (claude-haiku-4-5-20251001)	`platform.claude.com/docs`
claude-haiku-4-5	lifecycle	active	`platform.claude.com/docs`
claude-haiku-4-5	context	200,000 tokens	`platform.claude.com/docs`
claude-haiku-4-5	output	64,000 tokens max output	`platform.claude.com/docs`
claude-haiku-4-5	modality	text-in, image-in, text-out	`platform.claude.com/docs`
claude-haiku-4-5	pricing	1M input tokens: $1 · 1M output tokens: $5 · 1M cache write tokens (5m): $1.25 · 1M cache write tokens (1h): $2 · 1M cache read tokens: $0.1 · 1M batch input tokens: $0.5 · 1M batch output tokens: $2.5 (reference, not live quote)	`platform.claude.com/docs`
claude-haiku-4-5	sources	2 primary-source citations	`platform.claude.com/docsplatform.claude.com/docs`
claude-haiku-4-5	freshness	Fresh (last checked 2026-05-20)	—
claude-opus-4-7	identity	Anthropic (claude-opus-4-7)	`platform.claude.com/docs`
claude-opus-4-7	lifecycle	active	`platform.claude.com/docs`
claude-opus-4-7	context	1,000,000 tokens	`platform.claude.com/docs`
claude-opus-4-7	output	128,000 tokens max output	`platform.claude.com/docs`
claude-opus-4-7	modality	text-in, image-in, text-out	`platform.claude.com/docs`
claude-opus-4-7	pricing	1M input tokens: $5 · 1M output tokens: $25 · 1M cache write tokens (5m): $6.25 · 1M cache write tokens (1h): $10 · 1M cache read tokens: $0.5 · 1M batch input tokens: $2.5 · 1M batch output tokens: $12.5 (reference, not live quote)	`platform.claude.com/docs`
claude-opus-4-7	sources	2 primary-source citations	`platform.claude.com/docsplatform.claude.com/docs`
claude-opus-4-7	freshness	Fresh (last checked 2026-05-20)	—
claude-sonnet-4-6	identity	Anthropic (claude-sonnet-4-6)	`platform.claude.com/docs`
claude-sonnet-4-6	lifecycle	active	`platform.claude.com/docs`
claude-sonnet-4-6	context	1,000,000 tokens	`platform.claude.com/docs`
claude-sonnet-4-6	output	64,000 tokens max output	`platform.claude.com/docs`
claude-sonnet-4-6	modality	text-in, image-in, text-out	`platform.claude.com/docs`
claude-sonnet-4-6	pricing	1M input tokens: $3 · 1M output tokens: $15 · 1M cache write tokens (5m): $3.75 · 1M cache write tokens (1h): $6 · 1M cache read tokens: $0.3 · 1M batch input tokens: $1.5 · 1M batch output tokens: $7.5 (reference, not live quote)	`platform.claude.com/docs`
claude-sonnet-4-6	sources	2 primary-source citations	`platform.claude.com/docsplatform.claude.com/docs`
claude-sonnet-4-6	freshness	Fresh (last checked 2026-05-20)	—

Honest gaps

Data gaps (4)

Unverified canonical fields per model. Gaps are not invented; the catalogue refuses to guess.

Model	Field	Reason	Affected route
gemini-2-5-pro	release date	Vendor docs not yet retrieved or value not published.	/models/gemini-2-5-pro
claude-haiku-4-5	release date	Vendor docs not yet retrieved or value not published.	/models/claude-haiku-4-5
claude-opus-4-7	release date	Vendor docs not yet retrieved or value not published.	/models/claude-opus-4-7
claude-sonnet-4-6	release date	Vendor docs not yet retrieved or value not published.	/models/claude-sonnet-4-6

Sources

Source trail (3)

Primary-source citations referenced by the evidence rows above.

ai.google.dev/gemini-api
Google AI — Gemini 2.5 Pro model docs
official-vendor-docs · retrieved 2026-05-20
ai.google.dev/pricing
Google AI — Gemini pricing
official-vendor-pricing · retrieved 2026-05-20
platform.claude.com/docs
Anthropic — Models overview
official-vendor-docs · retrieved 2026-05-20

Outside the catalogue

Next external tests

The catalogue stops at verified fields. These are the checks a reader still needs to run before committing.

Run task-specific prompt tests against the candidate models in your own environment.
Verify request latency from your target deployment region.
Check rate limits in your provider account against your expected load.
Validate per-token cost against the vendor's current pricing page (references in this brief are not live quotes).
Confirm compliance / security requirements against your organisation's controls — verification status here is not a certification.

Export

Export this brief

Same query parameters as this page. Endpoint returns Markdown by default; pass ?format=json for the structured payload.

Build date: 2026-05-24. Briefs use this date as the deterministic generatedAt so the same build produces the same brief.