Claude Fable
A Claude AI model from Anthropic used for coding, planning and orchestrating agents.
Commenters like Fable as a planner and orchestrator that directs cheaper subagents such as Opus,1234 and for reliable code on hard problems.567 The most common complaint is that it is expensive and uses up limits quickly,891011 and some also dislike its writing style.1213
Why people recommend it
Planning and orchestration · 3
Many use it to plan and review while Opus, Sonnet or cheaper models do the implementation.
“I use Fable as the orchestrator, only speak with Fable”
“having fable plan and orchestrate with opus implemention + fable reviews, is my go-to.”
Show 1 more
“I now use Fable to create a PLAN.md.”
Code quality · 3
Commenters say its code is more reliable and careful than other models', with fewer janky results.
“With newer models, and especially large ones like Fable, it just doesn't happen anymore.”
“Fable 5.1 was noticeably better at getting things right and doing things in a good reliable way.”
Show 1 more
“Fable is the only one that fucking listens to me and writes proper Go code”
Hard problems · 3
Used for a math proof, unsolved ciphers, a Python interpreter in Rust and computational biology.
“I used Fable 5.1 and Astra to solve and prove a long-standing mathematical problem”
“Mostly solved with Fable 5.1 and Opus 5.”
Show 1 more
“I used Fable 5.1 and Opus 5.5 to build a Python interpreter in Rust. Appears to pass all the 3.12 code I throw at it.”
Code reviews · 3
Commenters praise the reviews it gives across specs and codebases.
“Fable 5.1 delivers incredible reviews.”
“I have Fable (Medium) review across the product spec, tech spec, and all the codebases”
Show 1 more
“having fable plan and orchestrate with opus implemention + fable reviews, is my go-to.”
PCB design · 3
Some report good results designing and routing circuit boards with KiCad tooling.
“I had Fable work on it, and it made a good board, after some trial and error.”
“I’ve had really good results recently with Fable 5.1.”
Show 1 more
“Fable took my Markdown notes, asked questions to fill in the blanks, and put together a plan.”
Caveats and dissent
Cost and limits · 3
Commenters say it is expensive and exhausts usage limits fast, sometimes on a separate meter.
“Anthropic has a separate meter for Fable. I used to get like five Fable sessions per week and that's it.”
Show 1 more
“In cost-per-task, it's a 32x multiple Fable vs DS41.”
Writing style · 3
Some find its prose and word choice poor and hard to steer.
“Fable 5.1 choice of words and general “words salad” expression are horrific.”
“The poor writing style made even models like Fable look worse than Sonnet 3.7 to me.”
Show 1 more
“Fable 5.1 was noticeably better at getting things right and doing things in a good reliable way.”
Perceived nerfing · 3
A few commenters believe it got worse after release.
“or fable before they lobotomised it”
“I stopped using Fable long time ago. It's worse than Sonnet.”
Show 1 more
“I have not tried Fable on it, because it was a car crash in previous similar task.”
Reasoning errors · 3
Some report it convincing itself of wrong conclusions or making basic mistakes.
“Opus and Fable always state something, then try to "prove" it but end up convincing themselves of the wrong thing.”
“LLMs can be unbelievably stupid. It recently happened a few times with Fable 5.1 as well.”
Show 1 more
“there’s concerning behavior for Fable on this front”
Unsupervised drift · 2
Commenters say it can over-engineer and cannot keep a codebase organized on long unsupervised tasks.
“Fable can venture into over-engineering itself into unnecessary complications.”
“not even fable can keep a codebase organized on long running unspervised tasks”
Often compared with
15 moved from Claude Opus to Claude Fable
5 moved from Claude Fable to Claude OpusOn the home page's Debates
1 moved from GPT Sol to Claude Fable
5 moved from Claude Fable to GPT Sol
3 moved from Claude Fable to CodexOn the home page's Debates
1 moved from GPT-5 to Claude Fable
2 moved from Claude Fable to GPT-5
1 moved from GPT Astra to Claude Fable
2 moved from Claude Fable to GPT Astra
Who recommends it in their own words
From the makers not counted in the ranking
“I think Fable 5.1 is a big improvement in writing style.”
Threads where it came up
- Claude Fable 5 35 mentions
- Claude Fable 5.1 and Claude Mythos 5.1 19 mentions
- Ask HN: What default model do you use and why? 11 mentions
- GPT-5.6 11 mentions
- US Government directive to suspend access to Fable 5 and Mythos 5 9 mentions
- Claude Opus 5.5 7 mentions
Also written as Fable 5, Claude Fable, Claude Fable 5, Fable 5.1. Data: claude-fable.json. Quotes are verbatim; the verdict and theme labels are written by a model from these comments, and every clause links to its sources.
All comments 48
“Fable is better at playing along with my "You are a dog" instruction in CLAUDE.md.”
“I don’t think it measures up to Fable 5.1 or Astra”
“Fable 5.1 choice of words and general “words salad” expression are horrific.”
“I only use Fable when "knowing about more stuff" is important (like research, exploring ideas)”
“or fable before they lobotomised it”
“where I use Fable and its limits spent slower than overall account limit”
“I used Fable 5.1 and Opus 5.5 to build a Python interpreter in Rust. Appears to pass all the 3.12 code I throw at it.”
“(max plan, haven't run out of the hourly or weekly limit once yet”
“fable scored 33%, while on the full challenge it landed closer to 14-15% range”
“"Distinguished Member of Technical Staff" -> Fable, plan mode.”
“I had Fable work on it, and it made a good board, after some trial and error.”
“I’ve had really good results recently with Fable 5.1.”
“With the new Fable 5.1, even the how part is getting replaced.”
“their new model B (Fable) capabilities”
“With newer models, and especially large ones like Fable, it just doesn't happen anymore.”
“occasional jumps up to Astra or Fable”
“We use Claude Fable to plan, and DeepSeek 4.1 Flash (hosted on DeepInfra) for everything else. Very cost effective.”
“We switched to Fable but it was a bit dissapointing”
“there’s concerning behavior for Fable on this front”
“Fable is great.”
“I use a mix of Fable 5.1, Opus 5.5, and Gemini 3.8 Flash”
“based on a formulation that Fable 5 originally proposed to me during a directed search”
“I tried out fable 5.1 the day it was released and coming from gpt-5.6-sol I was truly mind blown”
“fable was released and it seems my taste lines up with fable’s training quite a bit”
“I didn't experience issues when I used Fable via OpenRouter”
“he delegates complicated things to a project lead which is usually fable”
“even though I use Fable for making plans”
“If I were to switch to Anthropic, Opus 5.5 or Fable 5.1 on high or better.”
“I use plan mode constantly (with opus/fable)”
“that was a full 180 back to coherency”
“I use fable 5.1”
“I used Fable 5.1 and Astra to solve and prove a long-standing mathematical problem”
“i'm well aware of opus and fable's superior engineering qualities for writing code”
“but with Opus, Astra, or Fable it’s great”
“I used Fable some a bit back”
“Anthropic has a separate meter for Fable. I used to get like five Fable sessions per week and that's it.”
“Fable is on a class of its own when it comes to coding and orchestration”
“not even fable can keep a codebase organized on long running unspervised tasks”
“My main process is to use Fable to reason and then spawn Opus subagents, and I get amazing results”
“Also adding verification for that Fable 5.1 output in the same sesssion.”
“I use Fable as the orchestrator, only speak with Fable”
“The poor writing style made even models like Fable look worse than Sonnet 3.7 to me.”
“Fable chewed through my usage too quickly to be practical”
“having fable plan and orchestrate with opus implemention + fable reviews, is my go-to.”
“Fable 5.1 was noticeably better at getting things right and doing things in a good reliable way.”
“I don’t have a use for Fable at current prices.”
“Fable is the only one that fucking listens to me and writes proper Go code”
“I now use Fable to create a PLAN.md.”