Contains affiliate links. We may earn a commission — it never changes the verdict.Details
The Stack / Chat Assistants
Claude
02 DecisionsYes — Claude is the default reasoning engine in this stack. Sonnet 5 on the Pro plan covers 90% of operator work, and Claude Code turns it into a hands-on builder. If you only pay for one assistant, pay for this one.
The job in the stack
The thinking seat: long documents, hard reasoning, code, and anything where the words have to be right the first time.
Pricing
| Plan | Price | Notes |
|---|---|---|
| Free | $0 | Sonnet 5 as default model, daily limits |
| Pro | $20/mo | Higher limits, Claude Code, projects; Fable 5 metered via usage credits |
| Max | $100–200/mo | 5x–20x Pro limits for heavy daily use; Opus 5 is the default model here |
| Team | $25–30/user/mo | Central billing, collaboration features |
| API (Sonnet 5) | $2/M input | Now the standard price, not an intro rate — Anthropic cancelled the Sep 1, 2026 rise to $3/$15 on Aug 11, 2026. $10/M output |
| API (Opus 5) | $5/M input | $25/M output — near-frontier capability at half Fable 5's rate |
| API (Fable 5) | $10/M input | $50/M output — the frontier tier, priced like it |
Pricing verified 2026-08-22. Plans change — the source of truth is Claude's own pricing page.
Models & versions
- Claude Fable 5Frontier model, top of the Claude 5 family; metered by usage credits on paid plans since July 7, 2026
- Claude Mythos 5Limited-availability frontier model (anthropic.com/glasswing), priced identically to Fable 5 at $10/M in, $50/M out. Not something you can just sign up for — treat it as a preview, not a seat
- Claude Opus 5Near-Fable-5 capability at half the API price; 1M-token context in, 128K out; default model on Max and the strongest model available on Pro. Fast mode (research preview) buys quicker output at Fable 5 rates — $10/M in, $50/M out
- Claude Sonnet 5Default model on Free and Pro — the workhorse
- Claude Opus 4.8Superseded by Opus 5 on July 24, 2026; same API price, lower capability
- Claude Haiku 4.5Fast and cheap for high-volume API work
The honest read
What works
- +Best-in-class long-document reasoning and writing quality
- +Claude Code is the strongest agentic coding tool on the market
- +Sonnet 5 intro API pricing ($2/M in) undercuts rivals through August 2026
- +Projects and memory make it a genuine work environment, not a chat toy
What breaks
- −Fable 5 access is metered even on paid plans — heavy use gets expensive
- −No native image generation — you still need an image seat in the stack
- −Usage limits on Pro can pinch during long agentic sessions
- −Identity verification (government ID + live selfie, via Persona) can be triggered on some accounts for platform-integrity or age checks, effective July 8, 2026
Use it for
- →Operators who write for a living — scripts, briefs, newsletters
- →Builders using Claude Code as their engineering seat
- →Long-context work: contracts, research digests, codebases
Skip it if
- ×You mainly want image/video generation in one app — that's not this
- ×Your volume is light and free-tier ChatGPT or Gemini already covers it
Questions operators ask
?Is Claude Pro worth $20 a month?
If you write, plan, or code daily — yes. Sonnet 5 as the default model plus Claude Code access clears the bar on the first serious work session.
?What is the difference between Fable 5 and Sonnet 5?
Fable 5 is the frontier model — deepest reasoning, priced accordingly ($10/M input via API, metered on plans). Sonnet 5 is the fast workhorse and the default on Free and Pro. Start with Sonnet 5; escalate to Fable 5 only when a task actually needs it.
?Does Claude generate images?
No. Claude is text, reasoning, and code. Keep a separate image seat (Midjourney or Leonardo) in the stack.
What changed
- Freshness sweep, one change that matters: Sonnet 5's $2/$10 API pricing is permanent. Anthropic announced on Aug 11, 2026 that the increase to $3/$15 scheduled for Sep 1 will not happen — the launch rate is now the standard rate. If you costed a build assuming a September rise, that budget is now conservative. Also added: Claude Mythos 5, a limited-availability frontier model at Fable 5 prices, and Opus 5's fast mode (research preview, $10/$50). Consumer plans unchanged — Free, Pro $20, Max $100-200, Team $25-30/user.
- NEWS SCAN: Claude Opus 5 launched July 24, 2026 — 1M-token context in, 128K out, priced at $5/$25 per million tokens (the same rate Opus 4.8 held) and landing close to Fable 5 on capability at half Fable 5’s API price. It is now the default model on Max and the strongest model reachable on Pro. Opus 4.8 marked superseded; API pricing row added.
- NEWS SCAN: Anthropic launched Reflect (July 9, beta) — a built-in usage dashboard for Free/Pro/Max users with Memory enabled, showing topic breakdowns and activity patterns over 1/3/6/12 months, plus quiet-hours and break-nudge controls. Draws from Memory's compressed pattern summary, not raw transcripts; incognito chats excluded. Not a pricing or plan change — noted as a feature update.
- NEWS SCAN: Anthropic's identity-verification policy took effect July 8 — select accounts may be asked for a government ID plus a live selfie (processed by Persona) as part of routine platform-integrity or age checks; not required for all users. Not a pricing change; added as a trust-and-privacy disclosure per house voice. Separately confirmed the June 12 export-control suspension of Fable 5/Mythos 5 was fully resolved by July 1 (global Fable 5 access restored) — already correctly reflected here, no other changes needed.
- Verified plans and API pricing. Fable 5 moves to usage credits July 7; Sonnet 5 intro pricing runs through Aug 31, 2026.
- Claude Fable 5 launched — first model of the Claude 5 family, above Opus in capability.
- Claude Sonnet 5 released and made the default model on Free and Pro plans.
Give Claude a job
The thinking seat: long documents, hard reasoning, code, and anything where the words have to be right the first time.
Try Claude →Alternatives with different jobs
ChatGPT
02 DecisionsThe generalist seat: quick answers, brainstorming, voice mode, and the widest plugin/app ecosystem when you need breadth over depth.
Google Gemini
02 DecisionsThe Google-side seat: massive context windows, Deep Research reports, and native reach into Gmail, Docs, and Drive.