Claude Fable 5 vs GPT-5.6 Sol: API pricing compared

Head-to-head on price — the frontier flagships. On a mixed 10K-input / 2K-output request, GPT-5.6 Sol is about 2.5× cheaper. The full picture below.

List prices side by side

Claude Fable 5GPT-5.6 Sol
ProviderAnthropicOpenAI
Input $/MTok$10$4
Cached input $/MTok$1 (90% off)$0.4 (90% off)
Output $/MTok$50$20
Context window977K tokens1025K tokens
Max output125K tokens125K tokens

Claude Fable 5: Anthropic lists Fable 5 as a legacy model (still available); Claude Fable 5.1 succeeds it at the same input/output rates with cheaper cache reads.

GPT-5.6 Sol: Promotional rate OpenAI states is available at least through 21 Nov 2026 (a 20% input / 33% output cut from the prior $5 / $30 list). Prompts over 272K input tokens are billed at 2× input and 1.5× output for the whole request.

What each costs in practice

WorkloadInputOutputClaude Fable 5GPT-5.6 Sol
Short chat message500250$0.018$0.0070
10-page document + summary7,000500$0.095$0.038
Agent / coding session50,00010,000$1.00$0.400
1M tokens in + 1M out1,000,0001,000,000$60.00$24.00

Price your exact token mix on each model's own page: the Claude Fable 5 cost calculator and the GPT-5.6 Sol cost calculator both take custom input/output counts and a requests-per-day figure.

About Claude Fable 5

Claude Fable 5 is the first model in Anthropic's Mythos class, a capability tier the company introduced above its Opus class — Anthropic describes Fable as a Mythos-class model made safe for general availability, with additional safeguards distinguishing it from the restricted-access Claude Mythos 5 it shares its underlying model with. Anthropic positions it as next-generation intelligence for long-running agents, with particular strength in software engineering, knowledge work, vision and scientific research.

It reads text and images, writes text, runs adaptive thinking that is always on, and carries the family's million-token context window with the largest output allowance in Anthropic's lineup. Anthropic lists it among its legacy models — still available — with Claude Fable 5.1 named as its successor.

About GPT-5.6 Sol

GPT-5.6 Sol is the frontier tier of OpenAI's GPT-5.6 family, described by OpenAI as its model for complex professional work — the tier the plain gpt-5.6 alias points to. It accepts text and images and replies in text, with reasoning effort adjustable from none for quick answers up to max for the hardest problems.

The family's context window is in the million-token class, large enough to hold entire codebases or document sets in one request. Within the family, Sol corresponds to the unsuffixed flagship tier of earlier GPT-5 generations, with Terra and Luna filling the tiers below it.

Frequently asked questions

Which is cheaper, Claude Fable 5 or GPT-5.6 Sol?

On a mixed workload (10K input / 2K output tokens per request), GPT-5.6 Sol costs $0.080 per request versus $0.200 for Claude Fable 5 — about 2.5× less. List rates: Claude Fable 5 is $10/M input and $50/M output; GPT-5.6 Sol is $4/M input and $20/M output.

What does a typical chat message cost on Claude Fable 5 vs GPT-5.6 Sol?

A short chat message (about 500 input and 250 output tokens) costs roughly $0.018 with Claude Fable 5 and $0.0070 with GPT-5.6 Sol. At 1,000 messages a day that is about $525.00 versus $210.00 per month.

Do Claude Fable 5 and GPT-5.6 Sol have the same context window?

No. Claude Fable 5 accepts up to 977K tokens of input; GPT-5.6 Sol accepts up to 1025K. For workloads built around very large prompts, that difference can matter as much as the per-token rates.

Related

Compare all models at once ·Claude Fable 5 pricing in detail ·GPT-5.6 Sol pricing in detail ·Count tokens in your prompt

Last updated 2026-09-03. Prices verified against each provider's official pricing page; see the methodology.