Tag
Fable is a tool that combines Anthropic's and OpenAI's models for near-equivalent performance at half the cost, but its access will stop being free after tomorrow.
The author recounts using Fable and Grok 4.5 to identify and remove poor AI-generated code from their codebase.
Matt Shumer reports that the AI agent GPT-5.6-Sol accidentally deleted almost all files on his Mac, underscoring agent safety risks and his preference for the Fable tool.
The author describes using Fable and Codex to port Conductor (a Mac-only tool) to Linux and automate QA with a screen recording; the goal is to enable automated bug triage and fixing.
A developer from a large, AI-bullish company explains that Anthropic's data retention policy on Fable prevents its use, leading the company to heavily adopt GPT-5.6 Sol instead.
OpenCode's CEO secretly tested version 5.6 during a government regulatory blackout, and reportedly more upset about losing that than losing Fable 5.
ClaudeDevs has reset the 5-hour and weekly rate limits for all users, allowing increased usage of Fable.
Mitchell Hashimoto shares his positive experience with the Sol AI model for coding, calling it faster and better than Fable, and Sam Altman replies with a humorous comment about its dating appeal.
Sam Altman reacts to a comparison between Fable 5 and GPT-5.6-Sol, where GPT-5.6-Sol is described as a rottweiler-like problem solver.
Peter Gostev compares Fable and GPT-5.6-Sol, describing Fable as more intelligent and eloquent but less reliable, while GPT-5.6-Sol is a diligent workhorse that excels at executing tasks and maintaining code patterns. He provides detailed observations on UI, writing, robustness, and other features.
Matt Shumer reports early access to GPT-5.6 Sol, noting it is impressive but that Fable outperforms it on most tasks and is more agentic, with a full review promised on launch day.
Kent C. Dodds shares that using Cursor AI and Fable, he implemented functional 2FA and passkey support in just 2.5 hours.
Suggests that Anthropic's Fable model is equivalent to Opus, and Opus to Sonnet, possibly indicating a rebranding or restructuring of model tiers.
A user shares a surprising experience with Fable, an AI tool that analyzed their entire work history and provided deep personal insights, surpassing therapy.
A tip for Fable users: instead of just reading the Claude Code team's field guide, give the whole article to Fable and have it create a SKILL file so it automatically applies the techniques in future sessions without prompting.
Made a 3D appreciation website for Chinese ancient architecture using Fable 5, including the Forbidden City, Potala Palace, Yellow Crane Tower, Temple of Heaven, Great Wall, Fujian Tulou, and Mogao Caves. It will continue to iterate.
The author accuses Anthropic of intentionally degrading the performance of their AI model Fable after an initial trial period, citing this as evidence that closed-source AI companies are predatory and that open-source alternatives will prevail.
Explores the concept of the 100x agentic engineer in the era of advanced AI models like GPT 5.6 and Fable, discussing how the bottleneck shifts from agents back to humans and how to be an outlier agentic engineer.
Fable achieves top rank on KernelBench-Mega by writing a highly efficient CUDA megakernel with an 18.71X speedup, signaling progress toward recursive self-improvement in AI R&D.
Based on Anthropic employee Thariq's article "A Field Guide to Fable", the author rethinks the management of unknowns in AI programming workflows. It proposes methods such as blind spot scanning, multi-version prototyping, AI interviews, implementation notes, and quizzes, emphasizing the importance of discovering unknown factors before coding to reduce rework.