What Does Claude Fable 5.1 Cost?
Claude Fable 5.1 costs $10 per million input tokens and $50 per million output tokens. Send one million tokens in and get one million back, and that single exchange costs $60, our arithmetic from those two list prices.
Fable 5.1 is Anthropic's top Mythos-class tier, released in September 2026. Every number here was checked against Anthropic's own pricing documentation on 16 September 2026, which matters more than usual: the model is weeks old, and most of what is written about it predates the launch.
Output costs five times input. That ratio is the most important fact for budgeting, and we return to it below.
AI Perks tracks $7.7M in credits across 194 companies, which is the other half of this arithmetic.
How Claude Fable 5.1 Compares to the Rest of the Claude Line
Fable 5.1 is exactly twice the price of Claude Opus 5, five times Claude Sonnet 5, and ten times Claude Haiku 4.5, on both input and output.
| Model | Input / 1M | Output / 1M | 1M in + 1M out | Vs Fable 5.1 |
|---|---|---|---|---|
| Claude Fable 5.1 | $10.00 | $50.00 | $60.00 | 1x |
| Claude Mythos 5.1 | $10.00 | $50.00 | $60.00 | 1x |
| Claude Opus 5 | $5.00 | $25.00 | $30.00 | 0.5x |
| Claude Sonnet 5 | $2.00 | $10.00 | $12.00 | 0.2x |
| Claude Haiku 4.5 | $1.00 | $5.00 | $6.00 | 0.1x |
The last two columns are ours, derived from the listed prices. Mythos 5.1 matches Fable at $10 and $50 but is limited availability, so for most teams Fable is the real ceiling.
Notice the structure. Every current Claude tier prices output at exactly five times input, and the tiers themselves step 1x, 2x, 5x, 10x from Haiku upward. It is a legible ladder: estimate any tier by multiplying the Haiku number.
Claude Fable 5.1 vs GPT-6 Astra and the Wider Market
Claude Fable 5.1 and OpenAI's gpt-6-astra are priced identically: $10 input, $50 output, per million tokens. The two flagship models from the two largest labs landed on the same number within weeks.
| Model | Input / 1M | Output / 1M | 1M in + 1M out |
|---|---|---|---|
| Claude Fable 5.1 | $10.00 | $50.00 | $60.00 |
| gpt-6-astra | $10.00 | $50.00 | $60.00 |
| gpt-6-astra, above 200k context | $20.00 | $75.00 | $95.00 |
| gpt-5.5 | $5.00 | $30.00 | $35.00 |
| Claude Opus 5 | $5.00 | $25.00 | $30.00 |
| gpt-5.6-sol | $4.00 | $20.00 | $24.00 |
| gpt-5.4 | $2.50 | $15.00 | $17.50 |
| gpt-5.6-terra | $2.00 | $12.00 | $14.00 |
| Claude Sonnet 5 | $2.00 | $10.00 | $12.00 |
| Claude Haiku 4.5 | $1.00 | $5.00 | $6.00 |
| Meta Muse Spark 1.1 to 1.3 | $1.25 | $4.25 | $5.50 |
| gpt-5.6-luna | $0.20 | $1.20 | $1.40 |
| Meta Muse Spark Contributor | $0.10 | $0.20 | $0.30 |
The combined column is derived from the listed prices.
Four things worth pulling out. OpenAI charges a long context premium and Anthropic has not published one. Above 200,000 tokens, gpt-6-astra doubles input to $20 and raises output to $75. Anthropic has published no equivalent tier for Fable 5.1 in the documentation we checked, so $10 and $50 are the only Fable rates on record.
The Meta Contributor tier is the cheapest number on the page for a reason. Muse Spark drops to $0.10 and $0.20 there, and the trade is that Meta trains on your prompts and completions. Muse Image is $0.01 per image. Meta runs no startup credit program for Muse, so that bill is paid in cash.
A dated warning. OpenAI's gpt-5 ($1.25/$10), gpt-5-mini ($0.25/$2) and gpt-5-nano ($0.05/$0.40) hit API shutdown on 10 December 2026. Do not build anything new on them.
Two gaps we will not guess at. xAI's Grok token prices are reported inconsistently across sources, so check the xAI console rather than any third party table, this one included. Google Gemini prices were not verified here and are deliberately absent rather than estimated.
The Sonnet 5 Price Rise That Was Cancelled
Anthropic had scheduled Claude Sonnet 5 to go from $2 and $10 up to $3 and $15 per million tokens on 1 September 2026, then cancelled the increase. This is the most under-reported pricing fact of the quarter.
Sonnet 5's $2 and $10 was originally announced as an introductory rate running through 31 August 2026. It is now simply the standard price. Anthropic's pricing page states it plainly: "The previously scheduled increase to $3/$15 per million input/output tokens on September 1, 2026 will not occur."
The practical effect is a 33% saving against the price you may have budgeted for, and it holds Sonnet 5 at exactly one fifth of Fable 5.1 rather than drifting to 0.3x.
The same stability shows one tier down. Claude Opus 5 launched on 24 July 2026 at $5 and $25, the same price as Opus 4.8. Two consecutive Anthropic releases without a price rise is not the industry default.
What Claude Fable 5.1 Costs at Real Monthly Volumes
At 10 million input and 2 million output tokens a month, Claude Fable 5.1 costs about $200, against $100 for Opus 5, $40 for Sonnet 5 and $20 for Haiku 4.5.
| Monthly volume | Fable 5.1 | Opus 5 | Sonnet 5 | Haiku 4.5 |
|---|---|---|---|---|
| 1M in / 200k out | $20 | $10 | $4 | $2 |
| 10M in / 2M out | $200 | $100 | $40 | $20 |
| 100M in / 20M out | $2,000 | $1,000 | $400 | $200 |
| 1B in / 200M out | $20,000 | $10,000 | $4,000 | $2,000 |
These are estimates from the list prices above, at full rate on every token, with no batch or caching discount assumed.
The ratio matters more than the absolute figures. At a 5:1 input to output ratio, half your bill is output tokens. Push past that, which long form generation and verbose agent loops do, and output becomes the overwhelming majority of the invoice.
Trimming a verbose system prompt saves at the $10 rate. Capping response length, or routing a chatty summarisation step to Haiku, saves at the $50 rate. AI Perks exists for the other lever: not paying the bill at all.
Which Claude Tier Should You Build On?
Pick by workload economics, not by tier name. We have not verified capability benchmarks for these models, so the honest guidance is to route by cost and confirm quality with your own evaluations.
Claude Haiku 4.5 at $1 and $5 is the default for high volume, low judgement work: classification, extraction, routing, tagging. At one tenth of Fable, a workload has to be genuinely hard before the upgrade pays.
Claude Sonnet 5 at $2 and $10 is the workhorse, and the cancelled price rise makes it the best value line on the whole table. Most production traffic that needs real reasoning belongs here.
Claude Opus 5 at $5 and $25 sits at half of Fable and has held its price across two releases. It is the natural stop for teams wanting headroom above Sonnet without doubling again.
Claude Fable 5.1 at $10 and $50 is for the fraction of requests where the answer's quality is worth ten Haiku calls. In practice that is an escalation tier inside a router, not a default.
Because every Claude tier holds the same 5x output multiple, switching tiers changes your bill by a clean multiplier and never changes its shape.
Frequently Asked Questions
How much does Claude Fable 5.1 cost per million tokens?
Claude Fable 5.1 costs $10 per million input tokens and $50 per million output tokens, verified against Anthropic's own pricing documentation on 16 September 2026. One million tokens in plus one million out works out to $60. Credits that offset that bill are tracked at AI Perks.
Is Claude Fable 5.1 more expensive than GPT-6 Astra?
No. They are priced identically at $10 per million input and $50 per million output. OpenAI publishes a higher rate for gpt-6-astra requests above 200,000 tokens, at $20 input and $75 output. Anthropic has published no equivalent long context tier for Fable 5.1, so at standard context the two are level.
Did Claude Sonnet 5 get more expensive in September 2026?
No. An increase to $3 and $15 per million input and output tokens was scheduled for 1 September 2026 and then cancelled. Anthropic's pricing page states the increase will not occur. Sonnet 5 stays at $2 and $10, which is one fifth of Claude Fable 5.1 on both sides.
What is the cheapest current Claude model?
Claude Haiku 4.5, at $1 per million input tokens and $5 per million output tokens. That is one tenth of Claude Fable 5.1 and one fifth of Claude Opus 5. Ten million input and two million output tokens costs roughly $20 a month at Haiku rates. AI Perks tracks credits across 194 companies.
Does Claude Fable 5.1 charge extra for long context?
Anthropic has not published a long context surcharge for Claude Fable 5.1 in the pricing documentation checked on 16 September 2026, so the standard $10 and $50 rates are what is on record. OpenAI does publish one for gpt-6-astra, charging $20 input and $75 output above 200,000 tokens.
Can free credits cover Claude Fable 5.1 usage?
Yes. Provider credits are denominated in dollars, so they offset list prices directly. At Fable 5.1 rates, $1,000 of credit covers roughly 100 million input tokens or 20 million output tokens. AI Perks tracks $7.7M in credits across 194 companies at getaiperks.com.
How to Cover the Fable 5.1 Bill With Credits
Flagship pricing is survivable when someone else funds it. At $60 per million tokens in and out, a workload that would cost a small team thousands a month becomes a rounding error against a credit balance.
The arithmetic is simple. Every $1,000 of provider credit buys 100 million Fable 5.1 input tokens or 20 million output tokens at list price, and considerably more if you route the easy work to Sonnet 5 or Haiku 4.5. That is why tier selection and credit sourcing are the same decision.
AI Perks tracks $7.7M in credits across 194 companies, covering model providers, cloud compute and the tools around them. One place, current terms.
Run the frontier model. Let someone else pay the $50.