# Claude Fable 5.2 ships: lower token price, but early users say per-task cost and effort settings need checking

*By Coding Agents Alpha Tracker • November 10, 2026*

Anthropic's Fable 5.2 launch dominated the period. Practitioners shared pricing math, effort-level guidance, an ownership prompt, and when to pick it over Opus 5.5. Also covered: Simon Willison's live-iteration build, the Pi team's issue-triage bot, and new remote-machine controls for agents.

## Fable 5.2: cheaper tokens, similar cost per task, and effort settings that changed

**The release.** Cursor added Claude Fable 5.2 right away.
- **CursorBench:** 56.9% at default effort, up from 49.2% for Fable 5.1. At max effort it scores 60%, the top result on that benchmark [^1].
- **Price:** $8/M input and $40/M output tokens, 20% less than 5.1 [^2].
- **Speed:** Addy Osmani says it generates output up to 30% faster than 5.1 [^3].

**Watch the per-task cost.** Osmani says Fable 5.2 "uses a lot of tokens". In their testing, a task costs roughly the same as on Fable 5.1 despite the lower price [^4]. Theo's numbers on the $200 Claude Ultra plan point the same way. He estimates the plan covers about $15,000 of Opus 5.5 a month but only about $3,000 of Fable. Fable is also capped at 50% of the weekly limit and "burns faster per $" [^5].

Theo's preferred measure is work done per week:
- Opus only: about 1,200 turns and 230 threads.
- Fable until its limit, then Opus: about 900 turns but slightly more threads. He says Fable is more likely to finish work in one prompt [^6].

He based this on more than 1.1 million real API requests [^7]. He warns the numbers may shift, because much of his usage so far has been benchmarking the new model [^6].

**Don't reuse your old effort setting.** Cursor says effort levels behave differently from 5.1. Start at the default and compare levels at [cursor.com/evals](https://cursor.com/evals) [^2]. Theo reports that on the code-only benchmarks in Slopalytics, xhigh and max score *worse* than high. Fable's overall 5-point lead also shrinks a lot on code benchmarks [^8][^9]. Alex Albert gives similar advice: start at medium or high and save xhigh for the hardest problems [^10].

**Prompt tip: tell it the code is its own.** Anthropic's Fable 5.2 prompting guide, quoted by Simon Willison, says the model adds to existing structure, even repeating code, unless told the code is its own. The suggested sentence is: `This codebase is yours: what it's for and how it's built are your call, and you'll keep building on it.` On small coding tasks, it merged repeated code in about half of tries with this sentence and under a tenth without it [^11].

**Alex Albert's workflow tips** [^10]:
- Before it builds anything, ask it what's wrong with your idea. It will push back.
- Say up front what needs your approval, such as deletions, and that the rest is its call.
- Give it a whole project rather than the next step, and let it run for hours.

He also runs one audit prompt on every project whenever a new model arrives: *"Read this entire project and find everything that's broken, inconsistent, inefficient, or no longer needed. Fix what you're sure about and list the rest for me to review."* [^12]

**Fable or Opus 5.5?** Most early advice says to keep Opus as the default.
- **Osmani:** start most coding work on Opus 5.5 and use Fable for ambitious, open-ended work [^3].
- **@TheWholeDaisy:** use Opus when you already know what you want, and Fable for creative tasks where you don't [^13].
- **Theo's 3D-game test:** Fable took 76 minutes and about $42. He rated it "not significantly better than Opus 5.5 (esp. for the price)" [^14]. The Opus version had a better gameplay loop, controls and performance, and he thinks the two models will work well together on 3D-game tasks [^15].

## Simon Willison: let the agent see and change the running app

To find a beep among unopened boxes, Willison built Beep Locator ([repo](https://github.com/simonw/beep-locator)). It located the source to within about 40 cm: a carbon monoxide detector that needed new batteries [^16].

His steps:
1. **Start from a spec.** In Claude Code Cloud on his iPhone (Opus 5.5, Extra effort), with an empty repo, his first prompt was "Write out a problem.md file describing the problem." Claude found and read a 2007 paper on acoustic ranging while writing the spec [^16].
2. **Ask for hot-reload, not just the feature.** He told it to "add a mechanism so you can send new code to the listeners which they will then run automatically", so it could keep experimenting while he did something else [^16].
3. **Iterate live.** On his laptop, Claude Code with Fable 5.2 started the server, could see what was happening, and could change the software as it ran. An `installModule()` hook pushed code to the listening devices over WebSockets. That avoided page reloads, which on some devices re-trigger microphone permission prompts [^16].

## Handing routine work to agents

**Pi issue triage.** The Pi team was sorting 30–100 issues a day by hand. They built an issue-triage bot on Pi Durable. Sub-agents trigger on each new issue, do a first analysis and a first fix, and a maintainer approves the diff [^17]. A small classification guide updates as humans sign off on processed issues. They say the bot now matches their taste using past examples, which probably wouldn't have worked six months ago [^17].

**Kent C. Dodds' state-machine prompt:** "Find all of the implicit state machines in my app… elevate… UX, DX, and agent experience (AX)… without breaking anything. Also, make state machines explicit." [^18]

**Romain Huet (OpenAI)** says skills or AGENTS.md files untouched for about four months are "probably far outdated" [^19].

## Running agents on other machines

- **Dots (OpenAI):** you can now connect several local computers to one dot and choose which handles a task. For example, it can build and test on a dedicated machine while you keep coding on your laptop [^20].
- **T3 Code:** press cmd+k and search "mcp" to set up a connector for any agent you want to control T3 Code from [^21]. Theo's tip if you hit Grok Bot usage limits: get the $100 Claude plan, install T3 Code, set up that connector, and tell Grok Bot to "call T3 Code with Opus 5.5 for any task where it is applicable" [^22].
- **Brendan Long's sandbox:** he runs Claude Code with all permission prompts disabled ([clawed-abode](https://github.com/brendanlong/clawed-abode)) [^23]. To contain the risk, agents run as an unprivileged user inside an unprivileged VM. They use rootless Podman instead of Docker, because Docker access easily gives root [^23]. They reach the GPU only through a queue over SSH into a gVisor + nvproxy container. Its allowlist of NVIDIA driver calls reduces exposure to driver exploits, though imperfectly [^23].

---

### Sources

[^1]: [𝕏 post by @cursor_ai](https://x.com/cursor_ai/status/2119880829212172509)
[^2]: [𝕏 post by @cursor_ai](https://x.com/cursor_ai/status/2119880841728000057)
[^3]: [𝕏 post by @addyosmani](https://x.com/addyosmani/status/2119869672573251887)
[^4]: [𝕏 post by @addyosmani](https://x.com/addyosmani/status/2119907165007945813)
[^5]: [𝕏 post by @theo](https://x.com/theo/status/2119960608816988596)
[^6]: [𝕏 post by @theo](https://x.com/theo/status/2119962555326345641)
[^7]: [𝕏 post by @theo](https://x.com/theo/status/2119962809333346662)
[^8]: [𝕏 post by @theo](https://x.com/theo/status/2119946060361818578)
[^9]: [𝕏 post by @theo](https://x.com/theo/status/2119946288347402450)
[^10]: [𝕏 post by @alexalbert__](https://x.com/alexalbert__/status/2119869724247077229)
[^11]: [Quoting Prompting Claude Fable 5.2](https://simonwillison.net/2026/Nov/10/prompting-claude-fable-52)
[^12]: [𝕏 post by @alexalbert__](https://x.com/alexalbert__/status/2119884951550386671)
[^13]: [𝕏 post by @TheWholeDaisy](https://x.com/TheWholeDaisy/status/2119905339994960312)
[^14]: [𝕏 post by @theo](https://x.com/theo/status/2119926277801144695)
[^15]: [𝕏 post by @theo](https://x.com/theo/status/2119929646527717822)
[^16]: [Locating the source of a mysterious beep with Claude](https://simonwillison.net/2026/Nov/9/mysterious-beeps)
[^17]: [Ep 08 - "Nobody's Going to Write the Next React"](https://www.youtube.com/watch?v=kODgReWwrWM)
[^18]: [𝕏 post by @kentcdodds](https://x.com/kentcdodds/status/2119812466037428392)
[^19]: [Deep Dive 220 – Developer Experience at OpenAI: Romain Huet on building for agents](https://www.youtube.com/watch?v=0OjTLzteaZU)
[^20]: [𝕏 post by @OpenAIDevs](https://x.com/OpenAIDevs/status/2119883176726409645)
[^21]: [𝕏 post by @theo](https://x.com/theo/status/2119948325705462260)
[^22]: [𝕏 post by @theo](https://x.com/theo/status/2119994608516276256)
[^23]: [My Paranoid Coding Machine Setup](https://www.brendanlong.com/my-paranoid-coding-machine-setup.html)