Model comparison
Claude Fable 5 to Claude Fable 5.1: what the upgrade saves
Claude Fable 5.1 keeps Claude Fable 5's rates but cuts a cache hit from $1 to $0.25 per million. What that saves on agentic sessions, and what it doesn't.
· Prices as of September 26, 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 comparisonsClaude Fable 5
Anthropic · Released June 9, 2026 · Previous generation
The first Fable model, bringing Mythos-class capability to general use. Anthropic now recommends moving to Claude Fable 5.1.
Claude Fable 5 facts and comparisons
The short answer
Claude Fable 5.1 costs the same as Claude Fable 5 for input, output, and cache writes, but a cache hit drops from $1 to $0.25 per million tokens. That makes the example agentic coding session $10.50 instead of $12.00, 13% less, while uncached work costs the same on both. Anthropic recommends moving to Fable 5.1, and Fable 5 stays available as a legacy model.
Choose Claude Fable 5.1 if
- Your sessions read heavily from the cache, where each hit costs 75% less than on Fable 5.
- You use the fable alias in Claude Code, which now points to Fable 5.1.
- You want the Fable model Anthropic now recommends, which it calls its most capable generally available model.
Choose Claude Fable 5 if
- You have a pipeline validated on Fable 5 and want to move on your own schedule, since Anthropic keeps it available.
- Your workload is mostly uncached one-off requests, so the upgrade saves nothing on price and there is no cost pressure to switch.
Side by side
Specs and prices
| Fact | Claude Fable 5.1 | Claude Fable 5 |
|---|---|---|
| Maker | Anthropic | Anthropic |
| API model id | claude-fable-5-1 | claude-fable-5 |
| Released | September 1, 2026 | June 9, 2026 |
| Status | Current | Previous generation |
| Context window | 1M tokens | 1M tokens |
| Max output | 128K tokens | 128K tokens |
| Open weights | No | No |
| Input, per 1M tokens | $10 | $10 |
| Cache hit, per 1M | $0.25 | $1 |
| Cache write, per 1M | $12.50 (5-minute), $20 (1-hour) | $12.50 (5-minute), $20 (1-hour) |
| Output, per 1M tokens | $50 | $50 |
| Runs in | Claude Code, Cursor, OpenCode, OpenRouter, and GitHub Copilot | Claude Code, Cursor, OpenCode, OpenRouter, and GitHub Copilot |
Standard API rates in US dollars, as published by each maker on September 26, 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. Claude Fable 5: The full 1M context window is billed at standard rates.
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 | Claude Fable 5 |
|---|---|---|
| Agentic coding session, 100K input, 400K written to cache, 2M read from cache, 50K output | $10.50 | $12.00 |
| Large one-off review, 150K input with no cache hits, 10K output | $2.00 | $2.00 |
| Output-heavy generation, 30K input, 80K output | $4.30 | $4.30 |
| A month of sessions, 110 sessions: 5 a day, 22 working days | $1,155.00 | $1,320.00 |
| Where the session’s cost goes | ||
| Cache writes | $6.50 | $6.50 |
| Cache reads | $0.50 | $2.00 |
| Uncached input | $1.00 | $1.00 |
| Output | $2.50 | $2.50 |
- caching saves on the session with Claude Fable 5.1 (62%)
- $17.00
- caching saves on the session with Claude Fable 5 (56%)
- $15.50
What changed in price from Claude Fable 5 to 5.1?
One line of the rate card moved. Claude Fable 5.1 and Claude Fable 5 both charge $10 per million input tokens, $50 per million output tokens, and $12.50 and $20 for 5-minute and 1-hour cache writes. The cache hit fell from $1 to $0.25 per million, because Anthropic bills Fable 5.1 hits at 0.025x the input price instead of 0.1x.
Anthropic describes that as cache reads 75% cheaper, which it estimates makes typical workloads about 25% cheaper. The example agentic coding session, which writes 400K tokens to the cache and reads 2M, comes out 13% cheaper: $10.50 against $12.00. The saving is exactly the $1.50 less it costs to read those 2M cached tokens.
Why not 25%? In this session the unchanged cache writes are the largest cost, $6.50 on both models. Reads were 17% of the Fable 5 session and are 5% of the Fable 5.1 session. A workload that reads more from the cache for every token it writes saves a larger share, closer to Anthropic's estimate, and one that never hits the cache saves nothing: the large one-off review costs $2.00 and the output-heavy generation $4.30 on both.
What Anthropic says about Fable 5 and Fable 5.1
Anthropic introduced Fable 5 in June 2026 as the first Fable model, bringing Mythos-class capability to general use, and said: "Fable 5's capabilities exceed those of any model we've ever made generally available." It credited Fable 5 with working autonomously for longer than any previous Claude model and with staying focused across millions of tokens in long-running tasks.
Fable 5.1, released on September 1, 2026, is now Anthropic's most capable generally available model. Anthropic aims it at demanding reasoning and long-horizon agentic coding, lists long-running agentic coding and multistep research among its strengths, and suggests it when Opus-tier results fall short. For Fable 5 users, Anthropic's advice is to move to Fable 5.1.
Do you need to change anything to switch?
Almost nothing. Both models use Anthropic's newer tokenizer, which counts about 30% more tokens than earlier Claude models for the same text, so a prompt is the same number of tokens on either and the per-token comparison above holds directly. Both take 1M tokens of context at standard rates and write up to 128K of output.
Both also require 30-day data retention, so neither is available under zero data retention. In Claude Code the fable alias already points to Fable 5.1, so anyone choosing it with /model fable is on the new model. Fable 5 remains available as a legacy model for anyone who calls it by name.
One cost comes with switching mid-task: each model has its own cache, so moving an open session to Fable 5.1 writes your context again at $12.50 or $20 per million. Switching at the start of a new session avoids paying for that write twice. After that, every session like the example costs $1.50 less, which is $165.00 over 110 sessions a month at API rates.
Prompt caching
How Anthropic 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
Your own numbers
See what Claude Fable 5.1 and Claude Fable 5 really cost you.
everyaitoken reads your Claude Code, Cursor, OpenCode, and OpenRouter history on your Mac and prices every request at API rates, with what caching saved or cost. $9 once.
FAQ
Questions
Is Claude Fable 5.1 cheaper than Claude Fable 5?
Only on cache hits, which cost $0.25 per million instead of $1. Input, output, and cache writes cost the same. The example session costs $10.50 on Fable 5.1 and $12.00 on Fable 5.
Is Claude Fable 5 still available?
Yes. Anthropic lists it as a legacy model that stays available and recommends moving to Fable 5.1. In Claude Code, the fable alias now selects Fable 5.1.
Does Claude Fable 5.1 count tokens differently from Claude Fable 5?
No. Both use Anthropic's newer tokenizer, so the same text is the same number of tokens on either. That makes the price comparison a direct one.
How do I check what the cheaper cache hits save me?
EveryToken reads your Claude Code history on your Mac and prices every request at Anthropic's rates, with cache reads and writes shown separately. A week on each model side by side shows what the lower hit price is worth for your work.
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
- Anthropic docs: Claude Fable 5
- Anthropic: Claude Fable 5 and Claude Mythos 5
- Cursor docs: Claude Fable 5
- OpenRouter: Claude Fable 5
- Anthropic: Prompt caching