The Senior Colleague Who Tells You When It Is Not Sure
Claude has moved beyond careful chat. Fable 5 is still the most capable model Anthropic has released widely; Opus 5, launched July 24, is the new default on Claude Max and sits near that ceiling at half the price; Sonnet 5 holds the Free and Pro default. Opus 4.8 is now labeled legacy but still answers most flagged requests. Code, Cowork, Tag, Design and Microsoft 365 integrations put Claude directly inside the places work happens. Two dates are worth acting on this week: claude-opus-4-1-20250805 retires on August 5, 2026, and Sonnet 5 introductory pricing ends on August 31, 2026.
The colleague who will not pretend the work is done
Claude’s most useful personality trait is judgment. It is unusually willing to identify missing evidence, challenge the brief and keep uncertainty visible while still moving the task forward.
Careful without being passive
Claude reads deeply, preserves nuance and usually prefers an explicit caveat to a neat but unsupported conclusion.
- Strong source fidelity
- Structured, professional writing
- Useful self-review and error detection
Work where judgment matters
Long documents, complex coding, regulated analysis, technical reviews and deliverables that need a defensible reasoning trail.
- Board and client materials
- Repository-scale engineering
- Financial, legal and scientific workflows
The brief may itself be wrong
Claude is especially valuable when the input is incomplete, internally inconsistent or likely to push a model toward a polished fiction.
- Ask it to inspect the premise
- Invite direct disagreement
- Require uncertainty to remain visible
Analyze this proposal for the executive team. Before recommending anything, identify what is missing, internally inconsistent or too weakly evidenced to support a decision.
Create the strongest defensible synthesis from these sources. Separate: verified facts, reasonable inferences, unresolved questions and claims the material cannot support.
Opus 5 resets the middle of the range
The priced lineup runs from Haiku 4.5 at $1 / $5 per million tokens up to Fable 5 at $10 / $50, and the order is not what the names imply. Fable 5 is the most capable model Anthropic has released widely. Opus 5, released July 24, is the new default on Claude Max and the strongest model on Claude Pro. Sonnet 5 remains the default on Free and Pro. Opus 4.8 is labeled legacy in Anthropic’s own models overview yet still answers most flagged requests. Mythos 5 sits above all of them and is not generally available. Two dates dominate the next four weeks: Opus 4.1 retires on August 5, 2026 and Sonnet 5 introductory pricing ends on August 31, 2026.
| Model | Price per million tokens | Where it is the default | Status |
|---|---|---|---|
| Claude Fable 5 | $10 / $50 | Selectable on paid plans and the API | Most capable widely released model |
| Claude Opus 5 | $5 / $25 | Claude Max; strongest model offered on Claude Pro | Released July 24, 2026 |
| Claude Sonnet 5 | $2 / $10 now, $3 / $15 from September 1 | Claude Free and Claude Pro | Introductory pricing ends August 31, 2026 |
| Claude Haiku 4.5 | $1 / $5 | Not a plan default; the cheapest priced tier | Headline price re-verified August 3, 2026 |
| Claude Opus 4.8 | $5 / $25, unchanged by the July 24 release | Fallback for most flagged requests | Labeled legacy in the models overview |
| Claude Opus 4.1 | Price not re-verified for this edition | Wherever it is still pinned by API id | Retires August 5, 2026; Deprecated since June 5, 2026; replacement claude-opus-4-8 |
| Claude Mythos 5 | No price row of its own — the models overview says it shares Claude Fable 5’s specs and pricing | Not generally available | Gated to Project Glasswing |
Claude Opus 5 New Max default
Released July 24. Anthropic positions it as coming close to the frontier intelligence of Fable 5 at half the price, and places it within 0.5% of Fable 5’s peak score on the benchmark it cites. Treat both as vendor claims until independent results exist. It is now the default model on Claude Max and the strongest model available on Claude Pro.
- $5 / $25 per million tokens; 1M context, 128k maximum output
- Thinking is on by default; Anthropic states that fast mode is 2.5x faster at 2x base price — a vendor performance claim, time it on your own traffic
- In Claude Code, Claude Cowork and on Microsoft Foundry. Microsoft’s July 24, 2026 announcement also puts it in Microsoft 365 Copilot, but no Microsoft page names Claude Opus 5 in readable body text, so treat that as announced rather than documented — surface coverage unconfirmed
Claude Fable 5 Most capable, widely released
Still the strongest model Anthropic has made generally available. Anthropic reports it was state-of-the-art on nearly all tested benchmarks in software engineering, knowledge work, vision and scientific research at release — a vendor claim; verify on your own tasks. Anthropic also reported that Stripe completed a codebase-wide Ruby migration in a day that would have taken a team two months — a vendor case study, not an independent measurement.
- Available on every paid plan and the API; API id claude-fable-5
- A Covered Model: 30-day retention applies and it is not available under zero data retention
- The longer the task, the larger its lead
- Pricing unchanged at $10 / $50
Claude Sonnet 5 Free and Pro default
Anthropic’s most agentic Sonnet yet. It plans, uses browsers and terminals, and approaches Opus 4.8 performance on some tasks at a lower operating cost. The price it is being sold on today is not the price it will carry after August 31.
- Default for Free and Pro; available across all plans
- $2 / $10 introductory pricing reverts to $3 / $15 after August 31, 2026
- Budget renewals and volume forecasts on the higher number
Claude Opus 4.8 Legacy, still load-bearing
Anthropic’s models overview now labels Opus 4.8 legacy, but it has not left the stack. Flagged requests in Claude.ai, Claude Code and Claude Cowork still fall back to Opus 4.8 by default, so it remains the model that answers when a safeguard fires.
- Legacy label, active safety role
- Anthropic’s figure: four times less likely than Opus 4.7, the prior Opus generation, to let flaws in its own code pass — a vendor claim, verify on your own tasks
- Microsoft’s July 24, 2026 announcement says Opus 5 replaces it in the Microsoft 365 Copilot lineup; no Microsoft page names Claude Opus 5 in readable body text, so surface coverage is unconfirmed
Claude Mythos 5 Gated
The capability ceiling. Anthropic’s own Opus 5 announcement lists Mythos 5 above Fable 5 on capability, and it is still the same underlying model as Fable 5 with the cyber safeguards lifted. It remains restricted to Project Glasswing cyber defenders and critical infrastructure providers, and it continues to pick up the platform features Anthropic ships across the API as a whole.
- Not generally available — do not plan around it
- No printed price row: the models overview states that Mythos 5 shares Claude Fable 5’s specs and pricing, so the $10 / $50 figure is inherited by reference rather than quoted for Mythos 5 directly
- A Covered Model, like Fable 5: 30-day retention is mandatory and zero data retention is not offered
Where a blocked request goes Changed
The fallback picture is no longer uniform. Flagged requests in Claude.ai, Claude Code and Claude Cowork still default to Opus 4.8, but biology-related blocked requests now route to Opus 5. On the API, the fallbacks parameter accepts a "default" mode that applies Anthropic’s recommended fallback for each refusal category; server-side fallback is in beta. Those refusal categories rest on a published grading scheme: the Cyber Jailbreak Severity scale, which classifies how serious an attempted misuse is before a safeguard decides what to do about it.
- Anthropic reports that fewer than 5% of Fable 5 sessions trigger a safeguard fallback — a vendor figure, measure your own rate
- Test which model actually answers your flagged traffic
- The Cyber Jailbreak Severity scale, CJS-0 to CJS-4, was published July 2 with Amazon, Microsoft and Google
Claude for Teachers Free access
Verified US K-12 educators get free premium access. Sign-up is open through June 30, 2027.
- Announced July 14, 2026
- Verification required
- A recruiting channel as much as a product
Self-serve HIPAA Enterprise and Platform
HIPAA configuration became self-serve for Enterprise and Platform customers on July 14, removing a manual step from healthcare deployments.
- Configure without a bespoke process
- Still requires your own compliance review
- Confirm scope before moving regulated data
Claude Tag Team beta
A shared Claude teammate inside selected Slack channels, with channel-scoped context, tools, memories, budgets and audit logs.
- Team and Enterprise beta
- Shared channel continuity
- Administrative spend and permission controls
Claude Science Specialist workbench
A customizable scientific environment with tools, packages, compute and auditable artifacts.
- Research-oriented execution
- Transparent artifacts
- A sign of domain-specific workbenches to come
Dynamic workflows Preview
Claude Code can sustain larger jobs by changing how it works as the task unfolds rather than forcing every project through one fixed loop.
- Useful for long-running work
- Higher supervision requirement
- Treat preview behavior as non-deterministic
Mid-task instructions API
Developers can update system direction during a task without routing the change through a user turn or discarding useful prompt-cache structure. Mid-conversation system messages reached general availability on July 15.
- Change priorities while work runs
- Preserve clearer role boundaries
- Useful for orchestrated agent systems
Claude is now a family of work surfaces
Choosing Claude is no longer just choosing a model. The decisive question is where the work should run: conversation, terminal, governed desktop workspace, team channel, design surface or business application.
| Surface | Best for | Context advantage | Primary caution |
|---|---|---|---|
| Claude chat | Analysis, writing, research and files | Conversation, uploads, connectors and memory | Large context still needs coverage checks |
| Claude Code | Repository-scale engineering | Codebase, terminal, tools and project instructions | Command and prompt-injection risk |
| Claude Cowork | Desktop knowledge work | Selected folders, local apps, skills and plugins | Mount permissions and action review |
| Claude Tag | Shared team delegation | Channel history, scoped tools and shared memory | Spend, permissions and public mistakes |
| Claude Design | Rapid visual exploration | Brand inputs, conversation and prototypes | Still a Labs product; professional finishing required |
| Claude Science | Auditable scientific work | Packages, tools, compute and artifacts | Expert validation remains essential |
From source material to finished work without changing context
Anthropic’s direction is clear: Claude should follow the work across files, applications and teams instead of making users rebuild the brief in every surface.
Claude Code
Plans, edits, tests and debugs inside the real repository.
- Agent teams and subagents
- Long-running workflows
- Code review and verification
Claude Cowork
Runs multi-step tasks against selected desktop files and applications.
- Read-only or controlled write modes
- Reusable skills and plugins
- Dispatch from text or voice
Excel, PowerPoint and Word
Claude works directly in business documents and carries context between applications. Write tools shipped on July 7: Claude can draft and send email, act on the calendar and create or update files.
- Model, audit and explain spreadsheets
- Build decks from live analysis
- Review write permissions before granting them
Claude Tag
A shared team agent in Slack with scoped identity and memory.
- Channel-specific access
- Shared continuation
- Logs and budget controls
Claude Design
Explores prototypes and polished visual directions through conversation.
- Software experience ideation
- Slides and one-pagers
- Canva handoff
MCP and apps
Pulls live context from enterprise and creative systems.
- Google Drive and Gmail
- Market and portfolio data
- Blender and specialist tools
Skills and plugins
Packages instructions, tools and workflows into repeatable organizational capability.
- Reduce prompt folklore
- Standardize recurring work
- Deploy domain templates
Memory and context
Carries relevant context across sessions and, increasingly, across applications. On July 10 memory was reworked into individual categorized entries, replacing the daily memory summary.
- Preserve audience and standards
- Inspect and remove entries one at a time
- Review what should persist
Claude is becoming useful before, during and after the creative tool
Claude does not replace taste. Its advantage is expanding the number of directions a creative team can explore, automating repetitive production and keeping strategy connected to the artifact.
Claude Design Labs
Explore software experiences, prototypes, slides and one-pagers with visual iteration inside the conversation.
- Start with audience and desired behavior
- Include brand rules and references
- Export into professional finishing tools
Creative connectors MCP
Claude can work with tools such as Blender through connectors, scripts and the application’s own APIs.
- Analyze and debug scenes
- Batch repetitive changes
- Create custom tools inside the creative application
Cross-format production
Claude can turn research into a document, document into deck, data into charts and creative brief into a buildable prototype.
- Keep one narrative thread
- Reduce manual reformatting
- Use humans for taste and final selection
Build three distinct directions for this onboarding flow. Use our brand system, explain the behavioral idea behind each route, and make every prototype testable rather than decorative.
Inspect this Blender scene and identify the repetitive production steps. Propose a reversible automation plan, then implement only the steps I approve.
Turn this research and workbook into a 12-slide executive deck. Use the approved master, keep one claim per slide, and link every chart to its source.
Do not redesign this. Review it as a creative director: identify where the execution departs from the stated concept, audience or brand behavior.
Eight habits that make Claude substantially better
The model upgrade matters, but workplace behavior determines whether that capability becomes useful, expensive or risky.
Front-load the context
Give Claude the complete source pack and explain how the files relate before asking for conclusions.
Let it inspect the brief
Ask what is missing or inconsistent before it begins production.
Choose effort deliberately
Use lower effort for reversible work and escalate only when consequence justifies it.
Define completion
Specify the artifact, validation criteria and evidence report—not only the topic.
Keep context across stages
Move from analysis to file to deck without rebuilding the narrative in a new chat.
Use an adversarial second pass
Ask Claude to identify the weakest claims, broken assumptions and missing tests in its own work.
Name the audience and stakes
Claude changes depth, tone and caution when it knows who will use the result and what failure costs.
Trust the loop, inspect the result
Avoid micromanaging every step, but review commands, diffs, formulas and finished artifacts.
Produce the completed board memo—not an outline. Validate every recommendation against the attached evidence and include a one-page appendix of unresolved assumptions.
Add pagination to the users API, update the contract and documentation, add edge-case tests, run the full suite, and stop only when CI passes or you can name the blocker precisely.
Re-read your work as a skeptical domain expert. Identify the three conclusions most likely to be wrong and show what evidence would change each one.
Before synthesizing, create a coverage map of every supplied document. After the answer, list any sources or sections you did not materially use.
The more Claude can act, the more its environment matters
Agent capability changes the failure mode. A weak chat answer is inconvenient; a tool-using agent with broad file and network access can create real operational damage.
Prompt injection
Instructions inside email, documents, websites or pasted prompts can redirect an agent. Treat external content as untrusted input.
Mount and write access
Cowork permissions should match the task. Read-only, read-write and no-delete boundaries are meaningful controls.
Higher effort cost
More reasoning can improve hard work but adds latency, token use and the possibility of unnecessary exploration.
Preview instability
Dynamic workflows, Labs products and beta surfaces require regression testing before production use.
Long context omissions
A 1M-token window is capacity, not proof of complete recall or equal attention.
No native media generation
Claude can plan, analyze and prototype visuals, but final image, video and audio generation still depends on other tools.
Model-version drift
A model upgrade can change formatting, tool choices and reasoning style. Re-test critical prompts and workflows.
Diplomatic language
Claude may soften criticism unless directness is explicitly requested. Ask for the harshest defensible review.
August 5 model retirement
Per Anthropic’s deprecations page as read on August 3, 2026, claude-opus-4-1-20250805 retires on August 5, 2026 and claude-opus-4-8 is the named replacement. Anything still pinned to the 4.1 id — an integration, a saved evaluation, a scheduled job — stops working, and it will fail as a request error rather than a warning. It is the only Claude model retiring in August, September or October 2026, but it is not the only sunset in the window: the legacy Workbench and the three experimental /v1/experimental/*_prompt APIs retire on August 17, 2026. Book that migration in the same pass.
August 17 sunsets
The legacy Workbench and the three experimental /v1/experimental/*_prompt APIs retire on August 17, 2026 — two weeks from this edition’s August 3 verification date. Anything that still depends on them needs a migration path booked now.
July 24 breaking changes
Setting thinking to disabled is only valid at effort high or below; combining it with extra high or max returns a 400. Requesting fast speed on Opus 4.7 now errors. Both will break working integrations silently until a call fails.
Shifting defaults
Opus 5 became the Claude Max default on July 24 and Opus 4.8 is now labeled legacy. Pin a model in anything you evaluate, or your baseline changes underneath the test.
Vendor benchmark claims
The near-frontier positioning of Opus 5 and the 0.5% gap to Fable 5 come from Anthropic on Anthropic-cited benchmarks. Run your own task-level evaluation before moving production work down a price tier.
| Environment | Useful boundary | What still requires review |
|---|---|---|
| Claude chat | Ephemeral server-side container | Files produced, connector scope and claims |
| Claude Code | Local sandbox plus human command approvals | Shell commands, diffs, secrets and outbound access |
| Claude Cowork | Sealed VM and selected folder mounts | Mounted workspace, MCP servers and completed actions |
| Claude Tag | Channel-scoped tools, memory and spend | Public responses, tool actions and cross-team access |
First-party evidence behind this guide
Product status, availability and pricing can change quickly, and one price in this guide has a known expiry date. These are the primary Anthropic pages checked for this edition, verified August 3, 2026. The week to August 3 was quiet but not silent: Anthropic published one post, the Frontier Red Team’s July 30, 2026 write-up “Investigating three real-world incidents in our cybersecurity evaluations.” It is safety research with no pricing or model implications, it is not cited below because nothing in this guide rests on it, and it should not be read as a product change. One caveat about coverage: several operational dates in this guide — the July release-note items such as the Microsoft 365 write tools, the memory rework, self-serve HIPAA and mid-conversation system messages, plus the August 17 sunsets and the July 24 API breaking changes — come from Anthropic’s rolling release notes rather than a dated article, so they are not listed below, and a reader relying on any of them should confirm it directly.
AI Mindset