Claude Sonnet 5 and Fable 5: what actually changes for Claude Code users

Quick answer

Anthropic shipped Claude Sonnet 5 on June 30, 2026: a real complement to Opus 4.8 (and a credible alternative for most of your Claude Code sessions), with a 1M token context window and thinking that now adjusts itself. Claude Fable 5 had a very short life: generally available only since June 9, 2026, it was suspended for all customers on June 12, 2026, on order of the US government citing national security. As of July 1, 2026, Fable 5 (and its counterpart Mythos 5) remains unavailable, with no return date announced: the section dedicated to it below describes its characteristics for the record, not a model you can use today. Neither model replaces Opus or Haiku: both sit alongside the existing lineup.

Two launches, two philosophies

Picture your model lineup as a garage of mechanics. Haiku is the fast, cheap apprentice: great for oil changes and quick checks. Sonnet is the generalist mechanic you see almost every day: good on 90% of issues, reasonable on the bill. Opus is the senior expert you call for the tricky diagnostics. Fable is the manufacturer specialist you only see by appointment: very good at what it does, but it never shows you its full diagnostic notes (just a summary, if you ask), and it can decline a repair it considers unsafe.

Sonnet 5 touches your day-to-day usage directly: on coding and agentic tasks, it's a serious alternative to Opus 4.8 rather than a mere fallback model (details and numbers in the next section). Fable 5, already a more deliberate, occasional choice by design, has additionally had its doors shut since June 12, 2026 — see the callout above.

Claude Sonnet 5: the model you'll use most

What changed on the spec sheet

Sonnet 5 (API identifier claude-sonnet-5) costs $3 / $15 per million tokens (input/output), with an introductory rate of $2 / $10 through August 31, 2026. The context window grows to 1M tokens, with a maximum output of 128k tokens. Priority Tier is not available for this model yet.

Anthropic positions Sonnet 5 as the best combination of speed and intelligence across the whole Sonnet line, with quality close to Opus on coding and agentic tasks. On the cost-performance curves Anthropic published (agentic search on BrowseComp, computer use on OSWorld-Verified), Sonnet 5 at medium effort already shows substantially better cost-performance than Sonnet 4.6, and matches Opus 4.8 on some tasks at the highest effort levels. In practice: for most coding and agentic tasks that don't justify Opus 4.8's price tag, Sonnet 5 is now the alternative to try first, not just a cheaper fallback. You can see the detailed charts (BrowseComp, OSWorld-Verified, misaligned behavior) directly on Anthropic's announcement post (ouvre un nouvel onglet).

The real behavior change: adaptive thinking

Sonnet 5 turns on adaptive thinking by default: the model itself decides how much reasoning depth a task needs, without you configuring it. This is a behavior change compared to Sonnet 4.6.

The flip side: the manual budget_tokens parameter for extended thinking is gone. If your scripts, hooks, or SDK integration call the API directly with that parameter, you'll get an HTTP 400 error. Same goes for non-default sampling parameters like temperature, top_p, or top_k: any non-default value also returns a 400 error.

The new tokenizer trap

Sonnet 5 uses a new tokenizer that produces roughly 30% more tokens than Sonnet 4.6 for strictly identical text. In practice: your code, your prompts, and your file contents don't change, but their token "weight" goes up. Price per million tokens stays comparable to Sonnet 4.6, but since every exchange consumes more of them, your real-world costs and the available room in your context window need to be recalculated, not assumed stable.

Claude Fable 5: the specialist you call for the hard cases (suspended since June 12, 2026)

Fable 5 (API identifier claude-fable-5) was generally available from June 9 to June 12, 2026 on the Claude API, Claude Platform on AWS, Bedrock, Vertex AI, and Microsoft Foundry, before being suspended (see the callout above). Anthropic described it as its most capable widely released model to date.

It cost $10 / $50 per million tokens, with a default context window of 1M tokens and a maximum output of 128k tokens. That's more than double Sonnet 5's price, and twice as expensive as Opus 4.8 on both input and output.

Three documented behaviors (worth knowing if it's ever reactivated)

1. Thinking is always on. There's no "thinking off" mode for Fable 5: only the level adapts automatically to task complexity.
2. The raw reasoning chain is never exposed. You can request a summary via thinking.display: "summarized", but by default ("omitted"), you see nothing of the intermediate reasoning.
3. The model can refuse a request. Internal safety classifiers can trigger a refusal, returned as stop_reason: "refusal" in a 200 HTTP response (not an error). Server-side and client-side fallback mechanisms exist to handle this case.

Don't confuse Fable 5 with Mythos 5

Here's today's trap: Fable 5 is not the successor to Claude Mythos Preview. That role belongs to Claude Mythos 5, available in limited access through the Project Glasswing program — and also suspended since June 12, 2026, under the same government directive. Fable 5 and Mythos 5 share the same technical specs and the same price, but Mythos 5 doesn't carry Fable 5's safety classifiers. If you see a reference to "Mythos" in an announcement or docs, that's not the same product described here — and it isn't any more available today.

Comparison table: the current models

Sonnet 5's introductory rate ($2 / $10) expires on August 31, 2026. After that date, the standard rate ($3 / $15) applies automatically, no action needed on your end. Only Haiku 4.5, Sonnet 5, and Opus 4.8 are actually usable today: Fable 5 is listed here as a pricing reference, not as an available option.

What this actually changes in your Claude Code workflow

# One-off session
claude --model claude-sonnet-5

# Default for the whole shell
export ANTHROPIC_MODEL="claude-sonnet-5"

FAQ

Does Sonnet 5 become the new default model in Claude Code? That's not a verified fact at this point: public documentation as of June 30, 2026 describes the model's characteristics, not an imposed default change. Check your configuration (ANTHROPIC_MODEL, --model flag) before assuming an automatic switch.

Should I worry about the 30% token increase? Don't panic, but do recalculate. The introductory pricing offsets part of the increase through August 31, 2026. Measure with /cost instead of extrapolating.

Is Fable 5 directly accessible in Claude Code? No, and it isn't accessible anywhere else right now either. Fable 5 was suspended for all Anthropic customers on June 12, 2026, by US government order — including on the Claude API, AWS Bedrock, Google Vertex AI, and Microsoft Foundry. Targeting claude-fable-5 via ANTHROPIC_MODEL or --model will fail while the suspension is in effect, regardless of the provider configured.

Why was Fable 5 suspended? The US government cited an export-control directive tied to national security, after reportedly identifying a way to circumvent the model's safeguards. Anthropic disputes the proportionality of the measure but is complying with it. No reactivation date had been announced as of July 1, 2026.

Can I turn off Fable 5's thinking? While the model remains suspended, this question is moot. For the record, when it was accessible (June 9-12, 2026), thinking was always on for Fable 5, only its level adjusted automatically — a structural difference from the rest of the lineup.

Are Mythos 5 and Fable 5 the same thing? No. They share the same specs and price, but Mythos 5 (limited access via Project Glasswing) is the successor to Mythos Preview and doesn't carry the safety classifiers present in Fable 5. Both are covered by the same suspension since June 12, 2026.

Can Sonnet 5 replace Opus 4.8? Not in every case, but it's a genuine complement: on coding and agentic tasks, Sonnet 5 at high effort gets close to Opus 4.8's quality for a fraction of the cost. Reserve Opus 4.8 for tasks that truly justify its price (very complex reasoning, massive refactors) and try Sonnet 5 first on everything else.

Next steps

• Fable Is Back: What You Must Know: redeployment timeline, new safety guarantees, and access conditions since July 1
• The real costs of Claude Code: recalculate your budget with the new pricing
• Why your tokens cost more in some languages: understand the impact of a tokenizer change
• Plan Mode vs Thinking Mode: what adaptive thinking changes for your workflows
• Multi-provider: AWS Bedrock and Google Vertex AI: configure Claude Code for Sonnet 5 and Opus 4.8 on other providers
• CLI reference: every flag to switch models, including --model