Claude Fable 5 vs Claude Opus 5
Anthropic's two frontier tiers compared — Fable 5's top-end capability against Opus 5 at half the price.
Updated August 28, 2026 · Independent comparison — Claude Fable 5, Claude Opus 5 are separate products.
Claude Fable 5
Anthropic's most capable widely released model
Claude Fable 5 is Anthropic's most capable widely released model, aimed at the most demanding reasoning and long-horizon agentic work. It has a 1M-token context window, up to 128K output tokens, and adaptive thinking that is always on — the raw chain of thought is never returned, only optional summaries, with depth tuned through the effort ladder (default high). At $10/$50 per million tokens it is the priciest Claude — roughly twice Opus 5's cost — it requires a data-retention arrangement rather than zero-data-retention, and single hard turns can run many minutes.
Claude Opus 5
Anthropic's default for complex agentic coding
Claude Opus 5 is Anthropic's default for complex agentic coding and enterprise work — a step-change over Opus 4.8, released July 24, 2026, at half of Fable 5's price. It shares the 1M context and 128K output (up to 300K on the Batch API beta), keeps thinking on by default, and is the only tier here with a research-preview fast mode (up to 2.5x output speed at premium pricing). It costs $5/$25 per million tokens, with a $0.50 cache-read rate and cache-write at $6.25 (5-minute) or $10 (1-hour), and it runs on the Claude API, Bedrock, Vertex, and Microsoft Foundry.
Bottom line
Anthropic's two frontier tiers compared — Fable 5's top-end capability against Opus 5 at half the price. Choose Claude Fable 5 if you want the single most capable Claude tier and budget is secondary; choose Claude Opus 5 if you want frontier agentic coding at half the price — $5/$25 vs $10/$50.
Context & Limits
| Feature | Claude Fable 5 | Claude Opus 5 |
|---|---|---|
| Context window | 1M (1,000,000 tokens) | 1M (1,000,000 tokens) |
| Max output tokens | 128K | 128K (up to 300K on Batch API beta) |
| Image / vision input | ✓ | ✓ |
| Released | Newest Anthropic flagship | 2026-07-24 |
Pricing (per 1M tokens)
| Feature | Claude Fable 5 | Claude Opus 5 |
|---|---|---|
| Input | $10.00 | $5.00 |
| Output | $50.00 | $25.00 |
| Cached input read | $1.00 (0.1x input) | $0.50 |
| Batch API discount | 50% | 50% |
| Fast mode (premium speed) | ✗ | Up to 2.5x output speed (premium pricing) |
Coding & Agentic Standing
| Feature | Claude Fable 5 | Claude Opus 5 |
|---|---|---|
| Positioning | Anthropic's highest-capability tier for the hardest work | Anthropic's default for complex agentic coding; a step-change over Opus 4.8 |
| Relative capability | Top of the Claude line | Lands close to Fable 5 on most reasoning and agentic work, at half the price |
| Vendor benchmark suites reported | SWE-bench (Verified / Pro) | SWE-bench (Verified / Pro) |
| Test-time compute scaling | Strong | Strongest deep-reasoning scaling in the Opus line |
Access & Behavior
| Feature | Claude Fable 5 | Claude Opus 5 |
|---|---|---|
| Open weights | ✗ | ✗ |
| Thinking | Always on; cannot be disabled | On by default; disable allowed only at effort high or below |
| Effort ladder | low - max | low - max |
| Zero-data-retention option | No (retention arrangement required) | ✓ |
| Where you can run it | Claude API, Claude apps, Cursor Max Mode, OpenCode Zen, DevPass | Claude API, Bedrock, Vertex, Foundry; Claude Code on Pro/Max |
Speed & Fit
| Feature | Claude Fable 5 | Claude Opus 5 |
|---|---|---|
| Speed tier | Slowest, highest capability | Faster; optional fast mode up to 2.5x output tokens/sec |
| Best-fit workload | The single hardest reasoning and coding problems, budget aside | Complex agentic coding and enterprise work at production cost |
The Verdict
Choose Claude Fable 5 if...
- ✓You want the single most capable Claude tier and budget is secondary
- ✓Your work is the hardest reasoning and multi-hour autonomous runs
- ✓You need always-on maximum-depth thinking that can't be switched off
- ✓You're fine with a data-retention arrangement rather than zero-data-retention
Choose Claude Opus 5 if...
- ✓You want frontier agentic coding at half the price — $5/$25 vs $10/$50
- ✓You need zero-data-retention, which Fable 5 cannot offer
- ✓You want the option of fast mode for up to 2.5x output speed
- ✓You want frontier quality that lands close to Fable 5 on most work
- ✓You need Bedrock, Vertex, or Microsoft Foundry deployment, or Claude Code on Pro/Max
Frequently Asked Questions
What is the difference between Claude Fable 5 and Claude Opus 5?
Claude Fable 5 is Anthropic's most capable widely released model, aimed at the most demanding reasoning and long-horizon agentic work. Claude Opus 5 is Anthropic's default for complex agentic coding and enterprise work — a step-change over Opus 4.8, released July 24, 2026, at half of Fable 5's price.
How much do Claude Fable 5 and Claude Opus 5 cost?
Claude Fable 5: $10.00. Claude Opus 5: $5.00. See the pricing table above for full plan details.
Should I choose Claude Fable 5 or Claude Opus 5?
Choose Claude Fable 5 if you want the single most capable Claude tier and budget is secondary. Choose Claude Opus 5 if you want frontier agentic coding at half the price — $5/$25 vs $10/$50.
1DevToolDrive either Claude tier from one workspace
Fable 5 and Opus 5 are both closed Anthropic models you reach through Claude Code or the API. 1DevTool runs Claude Code alongside Codex and Gemini CLI in persistent terminals, so you can switch between Fable 5 and Opus 5 — or bring in a rival model — on your own keys or subscription, with no lock-in. It adds a built-in HTTP client, 13-engine database client, and embedded browser, all wired into Send-to-AI. One-time $29, no subscription.