Model comparison
Claude Fable 5.1 vs GPT-5.6 Sol: 2.5x apart, for now
GPT-5.6 Sol costs 60% less than Claude Fable 5.1 at promotional rates available at least through November 21, 2026. Why caching leaves the 2.5x gap intact.
· Prices as of September 28, 2026
Claude Fable 5.1
Anthropic · Released September 1, 2026
Anthropic's most capable generally available model, aimed at demanding reasoning and long-horizon agentic coding. Anthropic suggests it when Opus-tier results fall short.
Claude Fable 5.1 facts and comparisonsGPT-5.6 Sol
OpenAI · Released July 9, 2026 · Previous generation
The flagship of the July 2026 GPT-5.6 family, now succeeded by GPT-6 Sol and on promotional pricing.
GPT-5.6 Sol facts and comparisons
The short answer
Claude Fable 5.1 costs 2.5x as much as GPT-5.6 Sol on every list price, and the example agentic coding session keeps that ratio at $10.50 against $4.20. GPT-5.6 Sol is OpenAI's previous flagship, on promotional rates available at least through November 21, 2026, and Codex now suggests GPT-6 Sol in its place. Choose Fable 5.1 for Anthropic's most capable model in Claude Code, and GPT-5.6 Sol if you rely on it in Codex cloud or through the gpt-5.6 API id.
Choose Claude Fable 5.1 if
- You want Anthropic's most capable generally available model, which Anthropic suggests when Opus-tier results fall short.
- Your sessions reread large cached contexts: a Fable 5.1 hit costs $0.25 per million, less than GPT-5.6 Sol's $0.40.
- You send prompts above 272K input tokens, where GPT-5.6 Sol's long-context rates apply and Fable 5.1 stays at standard rates.
- Claude Code is where you work, and /model fable selects it there.
Choose GPT-5.6 Sol if
- You want the lower price while the promotion lasts: $4 input and $20 output per million tokens.
- You rely on Codex cloud chats through a ChatGPT plan, and those still run on GPT-5.6 Sol.
- Your work is frontend-heavy, where OpenAI claims better layout, visual hierarchy, and design judgment for GPT-5.6.
- You build agents with programmatic tool calling, in which the model writes JavaScript that calls tools and processes their output.
Side by side
Specs and prices
| Fact | Claude Fable 5.1 | GPT-5.6 Sol |
|---|---|---|
| Maker | Anthropic | OpenAI |
| API model id | claude-fable-5-1 | gpt-5.6-sol |
| Released | September 1, 2026 | July 9, 2026 |
| Status | Current | Previous generation |
| Context window | 1M tokens | 1.05M tokens |
| Max output | 128K tokens | 128K tokens |
| Open weights | No | No |
| Input, per 1M tokens | $10 | $4 |
| Cache hit, per 1M | $0.25 | $0.40 |
| Cache write, per 1M | $12.50 (5-minute), $20 (1-hour) | $5 |
| Output, per 1M tokens | $50 | $20 |
| Runs in | Claude Code, Cursor, OpenCode, OpenRouter, and GitHub Copilot | Codex, Cursor, OpenCode, OpenRouter, and GitHub Copilot |
Standard API rates in US dollars, as published by each maker (Claude Fable 5.1: September 26, 2026; GPT-5.6 Sol: September 28, 2026). Batch and priority tiers, taxes, and subscription plans are not included. Claude Fable 5.1: The full 1M context window is billed at standard rates. GPT-5.6 Sol: Requests over 272K input tokens cost 2x for input and cache and 1.5x for output, for the whole request. These are promotional rates, available at least through November 21, 2026.
Cost
What the same work costs
The same token counts, priced at each model’s published rates. Cached tokens are billed at each maker’s cache prices, so the session shows what caching is worth on each model.
Real sessions differ: the two models count the same code as different numbers of tokens, and reasoning settings change how much each one writes. Your own history is the real test.
| Workload | Claude Fable 5.1 | GPT-5.6 Sol |
|---|---|---|
| Agentic coding session, 100K input, 400K written to cache, 2M read from cache, 50K output | $10.50 | $4.20 |
| Large one-off review, 150K input with no cache hits, 10K output | $2.00 | $0.80 |
| Output-heavy generation, 30K input, 80K output | $4.30 | $1.72 |
| A month of sessions, 110 sessions: 5 a day, 22 working days | $1,155.00 | $462.00 |
| Where the session’s cost goes | ||
| Cache writes | $6.50 | $2.00 |
| Cache reads | $0.50 | $0.80 |
| Uncached input | $1.00 | $0.40 |
| Output | $2.50 | $1.00 |
- caching saves on the session with Claude Fable 5.1 (62%)
- $17.00
- caching saves on the session with GPT-5.6 Sol (62%)
- $6.80
Cheaper cache reads on the dearer model
Claude Fable 5.1 charges $10 input and $50 output per million tokens, 2.5x the $4 and $20 of GPT-5.6 Sol. Its cache hit breaks the pattern: Anthropic prices it at 0.025x input, $0.25, while OpenAI charges 0.1x on GPT-5.6 Sol, $0.40. Per cached token, the far pricier model is the cheaper one.
In the example session, 2M cached tokens cost $0.50 on Fable 5.1 and $0.80 on GPT-5.6 Sol. Writes more than make up the difference, $6.50 against $2.00, because Anthropic charges 2x input for its 1-hour cache and OpenAI 1.25x for every write. The session ends at $10.50 against $4.20, the same 2.5x as the uncached review, $2.00 against $0.80.
Caching saves $17.00 on Fable 5.1 and $6.80 on GPT-5.6 Sol against paying full input price for every token, 62% on each. Anthropic says Fable 5.1's cache reads are 75% cheaper than those of Claude Fable 5, which it estimates makes typical workloads about 25% cheaper.
How long will GPT-5.6 Sol's promotional pricing last?
GPT-5.6 Sol's $4 and $20 are promotional rates, which OpenAI says are available at least through November 21, 2026. The sources behind this page do not say what it will cost after that, so any comparison past that date needs a fresh look at OpenAI's pricing.
GPT-5.6 Sol is also a previous-generation model. It launched on July 9, 2026 as the flagship of the GPT-5.6 family, and OpenAI has since released GPT-6 Sol. Codex cloud chats on ChatGPT plans still use GPT-5.6 Sol, and the API id gpt-5.6 points to it, but elsewhere Codex suggests moving to GPT-6 Sol.
What Anthropic and OpenAI claim for each
Anthropic counts Fable 5.1 among "the world's most advanced models for coding and knowledge work" and names long-running agentic coding and multistep research as strengths. It runs adaptive thinking at high effort by default, and it requires 30-day data retention, so it is not available under zero data retention.
OpenAI describes GPT-5.6 Sol as its "Flagship model for complex professional work" and credits the GPT-5.6 family with token efficiency, reaching flagship-level performance with fewer output tokens. That claim, like reasoning settings and Anthropic's newer tokenizer, which counts about 30% more tokens than earlier Claude models, changes real token counts in ways a fixed-token table cannot show.
Both models take about a million tokens of context, 1M on Fable 5.1 and 1.05M on GPT-5.6 Sol, and both write up to 128K tokens. Above 272K input tokens OpenAI bills a GPT-5.6 Sol request at 2x for input and cache and 1.5x for output, for the whole request.
Prompt caching
How each maker bills cached tokens
Anthropic
Claude caches a prompt prefix up to a breakpoint. One top-level cache_control field places the breakpoint automatically and moves it as the conversation grows, or you can mark up to 4 blocks yourself. Claude Code manages caching for you.
A 5-minute cache write costs 1.25x the input price and a 1-hour write costs 2x. A cache hit costs 0.1x input on most models, 0.05x on Claude Opus 5.5, and 0.025x on Claude Fable 5.1 and Claude Mythos 5.1. Every hit restarts the cache lifetime at no charge.
Anthropic's rule of thumb: a 5-minute write pays for itself after one cache read, and a 1-hour write after two.
In Claude Code, the main conversation uses the 1-hour cache on a Claude subscription and the 5-minute cache with an API key. Each model has its own cache, so switching models starts over.
Source: Anthropic: Prompt caching
OpenAI
Prompt caching is on by default. From GPT-5.6 on you can also mark up to four explicit cache breakpoints, while GPT-5.5 and earlier cache automatically only.
From GPT-5.6 on, a cache write costs 1.25x the uncached input price and a cache hit costs 0.1x. GPT-5.5 and earlier add no charge for writing the cache: written tokens are billed as ordinary input, and a hit costs 0.1x on the models compared here.
On GPT-5.6 and later, a cached prefix stays reusable for at least 30 minutes after its last use, and caching starts at 1,024 input tokens.
Source: OpenAI: Prompt caching
Your own numbers
See what Claude Fable 5.1 and GPT-5.6 Sol really cost you.
everyaitoken reads your Claude Code, Cursor, OpenCode, OpenRouter, and Codex history on your Mac and prices every request at API rates, with what caching saved or cost. $9 once.
FAQ
Questions
How much more does Claude Fable 5.1 cost than GPT-5.6 Sol?
It costs 2.5x as much on input, output, and cache writes, but less on cache hits: $0.25 against $0.40 per million. The example agentic session costs $10.50 against $4.20, and 110 sessions a month $1,155.00 against $462.00, a $693.00 difference.
Are GPT-5.6 Sol's prices going up?
OpenAI calls its current rates promotional and lists them as available at least through November 21, 2026. What follows is not published in the sources behind this page.
Should I move from GPT-5.6 Sol to GPT-6 Sol?
Codex suggests that move everywhere except Codex cloud, where GPT-5.6 Sol still powers chats on ChatGPT plans. The API id gpt-5.6 still points to GPT-5.6 Sol, so code that calls it keeps working.
How can I compare their real costs?
EveryToken reads your local Claude Code, Codex, Cursor, and OpenCode history on a Mac and prices each request at API rates, split by model. It shows what caching saved or cost, which matters here because Fable 5.1's cheap reads and dear writes pull in opposite directions.
Sources
- Anthropic: Pricing
- Anthropic docs: Claude Fable 5.1
- Anthropic: Claude Fable 5.1 and Claude Mythos 5.1
- Claude Code docs: Model configuration
- Cursor docs: Claude Fable 5.1
- OpenRouter: Claude Fable 5.1
- OpenCode docs: Zen
- GitHub Docs: Supported AI models in Copilot
- OpenAI: API pricing
- OpenAI docs: GPT-5.6 Sol
- OpenAI: Using GPT-5.6
- OpenAI: API changelog
- Codex docs: Models
- Codex docs: Pricing
- Cursor docs: Models and pricing
- OpenRouter: GPT-5.6 Sol
- Anthropic: Prompt caching
- OpenAI: Prompt caching