Claude Fable 5.1 Is Here: Big Benchmark Gains, Lower Costs and What’s New
Written by
FABLE 5.1 DROPS ... AND ANTHROPIC BROUGHT THE RECEIPTS!
Bigger benchmark scores, cheaper cache reads, and a fresh contender for your toughest coding jobs.
Well, well, well ... look who just walked into the AI heavyweight fight.
Anthropic has released Claude Fable 5.1, alongside Claude Mythos 5.1, pitching its latest models at coding, research, and demanding knowledge work. And that innocent-looking “.1”? The benchmark jump gives it some serious attitude.
SHOW US THE NUMBERS!
Here are selected results from Anthropic’s published launch comparison. Higher percentages are better.
| Benchmark | Fable 5.1 | Fable 5 | Opus 5 | GPT-5.6 Sol |
|---|---|---|---|---|
| Terminal-Bench-Science 0.1 | 52.6% | 24.7% | 29.0% | 22.4% |
| Terminal-Bench 4.0 | 55.8% | 42.0% | 52.3% | 37.3% |
| AutomationBench | 31.4% | 17.1% | 26.9% | 19.6% |
| CursorBench 3.2.0 | 73.4% | 70.5% | 70.0% | 67.2% |
The attention-grabber: 52.6% on Terminal-Bench-Science, more than double Fable 5’s score. That is a jump on a specific scientific-research benchmark—not proof of double the intelligence across everything.
CURSOR HAS ITS OWN RECEIPTS
Cursor’s leaderboard also puts Fable 5.1 Max at 73.4%, ahead of Fable 5 Max at 70.5%. Its listed evaluation cost falls from $17.32 to $9.64. Those are benchmark costs, not a flat price for your next coding project.
THE PRICE TAG HAS A TWIST
API pricing stays at $10 per million input tokens and $50 per million output tokens. Cache reads drop to $0.25 per million tokens—75% below Fable 5’s rate.
Under the hood: a 1-million-token context window, up to 128,000 output tokens, and adjustable reasoning effort. Developers can access it using claude-fable-5-1.
BIG ENTRANCE. NOW COMES THE AUDITION.
These results depend on evaluation conditions, tools, and safeguards. Your messy codebase gets the next vote.
Fable 5.1 has made its entrance. Hand it the bug that ruined your weekend—and see whether it earns the applause.
Comments
No comments yet. Be the first to comment!
Leave a Comment