Claude Fable 5.2 Pricing: Official API Rates and Full Comparison (2026-11)
As of 2026-11-12, Claude Fable 5.2's official API price is $8 input, $40 output and $0.20 cache reads per million tokens, and Anthropic's pricing page and migration guide both put it below Fable 5.1's $10 / $50. Below, the official pricing page is listed item by item for 5.2, 5.1, Opus 5.5, Sonnet 5.5 and Haiku 5.5 input, output, cache writes and cache reads, with batch and fast mode prices, worked cost examples you can recalculate, and the official rules that change a bill.
Updated 2026-11-13
Prices change: on QCode, Claude, GPT and Chinese models are billed per token — live rates on /models.
- Billed per token — live rates on /models
- Pay by card (Visa / Mastercard), Apple Pay, Google Pay or crypto
- Self-service: top up and activate instantly
Official prices and cache rates
Fable 5.2 official price (input / output, USD)
Anthropic's official pricing page (checked 2026-11-12) gives Claude Fable 5.2 a base input price of $8 and output of $40 per million tokens; prompt caching is $10 for a 5-minute write, $16 for a 1-hour write, and $0.20 for cache hits and refreshes. An official footnote prices Fable 5.2 cache hits at 0.025x the base input price; the official migration guide lists Fable 5.1 at $10 / $50 with $0.25 cache reads.
Fable 5.1 official price (input / output)
The official pricing page's additional-models table lists Claude Fable 5.1: $10 base input, $50 output, $12.50 for a 5-minute cache write, $20 for a 1-hour cache write and $0.25 for cache hits, with cache hits at 0.025x the base input price. Against 5.2's $8 / $40, the 5.1 input, output and cache-read prices are all higher.
Opus 5.5 official price (input / output)
The official pricing page lists Claude Opus 5.5: $4 input, $20 output, $5 for a 5-minute cache write, $8 for a 1-hour cache write and $0.20 for cache hits; a footnote prices Opus 5.5 cache hits at 0.05x the base input price. The official Fable 5.2 overview comparison table also gives Opus 5.5 at $4 / $20, with a 1M context and 128K max output.
Sonnet 5.5 and Haiku 5.5 official price (input / output)
The official pricing page lists Claude Sonnet 5.5: $2 input, $10 output, $2.50 for a 5-minute cache write, $4 for a 1-hour cache write and $0.10 for cache hits (0.05x the base input price). Claude Haiku 5.5 has two tiers by prompt length: up to 100,000 tokens costs $0.10 input, $0.50 output and $0.01 cache hits; over 100,000 tokens costs $0.50 input, $2.50 output and $0.05 cache hits.
Item-by-item price comparison
As of 2026-11-12, Fable 5.2's official API price is $8 input, $40 output and $0.20 cache reads per million tokens. From the official Anthropic pricing page, item by item (per million tokens): Fable 5.2 $8 input, $40 output, $10 for a 5-minute cache write, $16 for a 1-hour cache write, $0.20 cache reads; Fable 5.1 $10 input, $50 output, $12.50 for a 5-minute cache write, $20 for a 1-hour cache write, $0.25 cache reads; Opus 5.5 $4 input, $20 output, $5 for a 5-minute cache write, $8 for a 1-hour cache write, $0.20 cache reads; Sonnet 5.5 $2 input, $10 output, $2.50 for a 5-minute cache write, $4 for a 1-hour cache write, $0.10 cache reads; Haiku 5.5 at prompts up to 100,000 tokens $0.10 input, $0.50 output, $0.01 cache reads, and over 100,000 tokens $0.50 input, $2.50 output, $0.05 cache reads. The same page lists Fable 5.2 batch pricing at $4 input and $20 output, and fast mode pricing at $16 input and $80 output.
Official rules that change the bill
The official pricing page and docs state the following rules that change a bill. First, cache writes and cache reads are charged at different points: cache write tokens are charged when content is first stored, cache read tokens when a later request retrieves the cached content; a 5-minute cache write is 1.25x the base input price, a 1-hour cache write is 2x, and a cache hit read is 0.1x on most models or 0.025x on Fable 5.2. Second, thinking tokens count as output: the official migration guide says thinking tokens are billed as output tokens, and max_tokens covers thinking plus response text. Third, long context: the official pricing page says 4.6 and later models (except Haiku 5.5) include the 1M context window at standard pricing, with a 900k-token request billed at the same per-token rate as a 9k-token request; Haiku 5.5 is priced by prompt length, and a request over 100,000 tokens pays higher prices. Fourth, cache multipliers stack with batch processing and data residency multipliers, and fast mode is not available with the Batch API.
Timeline
Anthropic's official model overview says Claude Fable 5.2 was released on 2026-11-09, with model ID claude-fable-5-2, a 1M context and 128K max output.
The Claude Code official changelog 2.1.321 says that from 2026-11-09 claude-fable-5-2 is the default Fable model, saved Fable 5.1 picks move to fable, and /model claude-fable-5-1 switches back.
This page checks the Anthropic pricing page, the Fable 5.2 overview and migration guide, and qcode.cc/models; official statements and QCode specs were all checked on 2026-11-12.
Confirmed vs not published
Confirmed (official wording)
First, the official pricing page says all prices are in USD and MTok means million tokens; Fable 5.2's official price is $8 input, $40 output and $0.20 cache reads (checked 2026-11-12). Second, the official model overview says Fable 5.2 was released on 2026-11-09, with model ID claude-fable-5-2, a 1M token context, 128K max output, adaptive thinking always on and a default effort of high. Third, the official migration guide says Fable 5.2 is priced below Fable 5.1's $10 / $50, with cache reads at $0.20 against $0.25, and that thinking tokens are billed as output tokens. Fourth, the official batch table lists Fable 5.2 at $4 input and $20 output, and the fast mode table at $16 input and $80 output; batch processing bills input and output tokens at half the standard price. Fifth, the official pricing page says 4.6 and later models (except Haiku 5.5) include the 1M context at standard pricing, and a Haiku 5.5 prompt over 100,000 tokens pays higher prices. Sixth, official footnotes price cache hits on Fable 5.2 and Fable 5.1 at 0.025x the base input price, and on Opus 5.5 and Sonnet 5.5 at 0.05x. Seventh, the Claude Code official changelog 2.1.321 says Fable 5.2 became the default Fable model on 2026-11-09.
Not published by the vendor
The official pricing page and model overview give no estimate of what a typical coding session costs on average; the cost examples here are conversions from the official unit prices with chosen input volumes, not a vendor-published average. The official overview also says Claude Mythos 5.2 is offered only to organizations verified through Anthropic's verification programs, such as the Cyber Verification Program, and shares Fable 5.2's specifications and pricing; the Fable model QCode offers is claude-fable-5-2.
How to choose
Against Fable 5.1
The official migration guide says Fable 5.2 is priced below Fable 5.1: $8 input against $10, $40 output against $50, and $0.20 cache reads against $0.25, with the same API surface and always-on adaptive thinking as 5.1. To work out a bill from the official prices, the 5.2 input, output and cache-read unit prices are all lower than 5.1; for a pipeline already running on 5.1, check the breaking changes listed in the official migration guide before switching. The migration guide also notes that at higher effort levels such as xhigh and max, Fable 5.2 was observed using subagents to review its own work, and subagents can account for a large part of what a session costs.
Against Opus 5.5
The official Fable 5.2 overview comparison table gives Opus 5.5 at $4 / $20 and Fable 5.2 at $8 / $40, both with a 1M context, 128K max output and $0.20 cache reads. The official What's new page for Fable 5.2 advises starting with the latest Opus model for most workloads and moving to Fable 5.2 only when evals on an Opus model at higher effort still fall short. The migration guide adds that Fable 5.2 defaults to high effort where Opus 5.5 defaults to medium, so a request that omits effort runs one level higher. At the official prices, the same 100,000-input, 10,000-output request costs $0.40 + $0.20 = $0.60 on Opus 5.5 against $1.20 on Fable 5.2 (same formula as Example 1 below). To estimate cost at the official prices, put either column of unit prices into the examples.
How to estimate the cost
Example 1 (converted from the official unit prices): 100,000 input tokens + 10,000 output tokens on Fable 5.2. Input cost = 100,000 ÷ 1,000,000 × $8 = $0.80; output cost = 10,000 ÷ 1,000,000 × $40 = $0.40; total $1.20. Example 2 (what cache reads change): 100,000 input tokens all billed at the base input price cost 100,000 ÷ 1,000,000 × $8 = $0.80; if 90% of the input is a cache hit, the input cost at the official $0.20 cache-read and $8 base-input prices is 90,000 ÷ 1,000,000 × $0.20 + 10,000 ÷ 1,000,000 × $8 = $0.018 + $0.08 = $0.098. Example 3 (batch pricing): the same 100,000 input + 10,000 output through batch processing, at the official Fable 5.2 batch prices of $4 input and $20 output; input cost = 100,000 ÷ 1,000,000 × $4 = $0.40, output cost = 10,000 ÷ 1,000,000 × $20 = $0.20, total $0.60. The input token counts in these formulas are chosen values; readers can substitute their own volumes. For release status see /claude-fable-5-2-status-tracker, for the older Fable 5 pricing see /claude-fable-5-pricing-guide, and for Opus 5.5 single-model specs see /claude-opus-5-5-guide.
How to use it on QCode
QCode supports claude-fable-5-2: in Claude Code and other Anthropic-protocol tools, set the model to claude-fable-5-2 with Base URL https://api.qcode.cc/api; the same key also calls claude-fable-5-1, claude-fable-5, claude-opus-5-5 and more. One key covers both Anthropic-compatible and OpenAI-compatible endpoints (Claude models use the Anthropic endpoint above). Billing is per token, with per-model rates on /models. Not ready for a monthly plan? Pay-as-you-go Packs start at $4 for a one-time quota you use within 7 days; if you already have a Personal Plan, add a booster pack when you need more quota.
Frequently asked questions
How much does Fable 5.2 cost?
As of 2026-11-12, the official API price is $8 input, $40 output and $0.20 cache reads per million tokens (Anthropic's official pricing page); Fable 5.1 is $10 / $50 with $0.25 cache reads. Batch pricing is $4 / $20 and fast mode pricing is $16 / $80.
How do the official prices compare with Fable 5.1?
The official pricing page and migration guide both say: Fable 5.2 $8 input against Fable 5.1's $10, $40 output against $50, and $0.20 cache reads against $0.25; a 5-minute cache write is $10 against $12.50, and a 1-hour cache write is $16 against $20.
How does it compare with Opus 5.5?
The official Fable 5.2 overview comparison table gives Opus 5.5 at $4 / $20 and Fable 5.2 at $8 / $40, both with a 1M context, 128K max output and $0.20 cache reads; the official migration guide also gives Fable 5.2's $8 / $40 against Opus 5.5's $4 / $20.
How does prompt caching affect the bill?
The official pricing page charges cache write tokens and cache read tokens at different points: cache write tokens are charged when content is first stored, cache read tokens when a later request retrieves it; a 5-minute cache write is 1.25x the base input price, a 1-hour cache write is 2x, and a cache hit read is 0.1x on most models or 0.025x on Fable 5.2. The official page also states these multipliers stack with batch processing and data residency multipliers.
Is Fable 5.2 included in a subscription?
Whether a subscription includes Fable 5.2 and how much each tier allows are set out on the official plan table at claude.com/pricing; subscription quota details have their own page.
How do I switch to Fable 5.2 in Claude Code?
Per the Claude Code official docs, use /model followed by a model name to switch immediately, or run /model with no argument to open the picker. For Fable 5.2, run /model fable or launch with claude --model fable; in Claude apps gateway sessions, or through a gateway configured to serve the model, select it with /model claude-fable-5-2. The official changelog 2.1.321 says Fable 5.2 is the default Fable model from 2026-11-09, and saved Fable 5.1 picks move to fable (at once on Pro and Max plans, on Team and Enterprise plans once the organization allows it); /model claude-fable-5-1 switches back.
Sources
Anthropic official sources: the pricing page, the Claude Fable 5.2 model overview, What's new and the migration guide; Claude Code official sources: the changelog and model-config docs; QCode: qcode.cc/models, qcode.cc/pay-as-you-go, qcode.cc/claude-fable-5-2-status-tracker and the docs.qcode.cc endpoints page. All were fetched and checked on 2026-11-12.
Estimate at official rates, pay per token
Not ready for a monthly plan? Pay-as-you-go Packs give you a one-time quota you use within 7 days; if you already have a Personal Plan, add a booster pack when you need more quota.
Related
One-time quota packs: no monthly plan, valid 7 days
One-time quota, no subscription, valid 7 days; per-model rates are on /models.
Claude Fable 5.2 release status
Fable 5.2 release date, model ID and availability.
Claude Fable 5.2 vs Claude Opus 5.5 (2026-11)
Official comparison of Fable 5.2 and Opus 5.5: release dates, positioning, context, max output, official prices, thinking behavior and API differences, plus a selection guide built on the official statements and Claude Code switch commands.
Crypto payment guide
Nine coins, 56 networks, zero platform fee for AI coding API credits.
Official prices and rules were checked on 2026-11-12 against Anthropic's official pages; model availability and per-model rates follow qcode.cc/models.