Anthropic LO ENTENDIÓ TODO con Fable 5.1

Value of the Information & Strength of the Argument

The video provides valuable, up-to-date information about a major AI model release, including pricing, benchmark data, and a real-world comparison. The host’s argumentation is solid: he uses a practical test to illustrate the models’ capabilities and limitations, and he acknowledges the limitations of self-reported benchmarks by referencing independent data. The comparison is fair, using the same prompt and conditions for both models. The host’s subjective preferences are clearly stated, and he provides concrete examples to support his opinions, such as the design choices made by each model. The video also offers context about the previous Fable 5 controversy, which helps viewers understand the significance of the new release.

Scientific Rigor, Source Quality, Title Accuracy

The video demonstrates scientific rigor by referencing independent benchmarks from Artificial Analysis and acknowledging the potential bias in Anthropic’s own benchmarks. The host also provides specific pricing details and explains the implications of cache read cost reductions. The title accurately reflects the content, focusing on Anthropic’s response to previous criticism. The video includes a practical test that, while not a controlled experiment, offers a transparent and reproducible methodology. The host’s commentary is balanced, noting both strengths and weaknesses of the models. The video does not cite external sources beyond the company’s own materials and the independent benchmark site, but it does not make unsupported claims.

The title accurately reflects the content: the video focuses on Anthropic's new Fable 5.1 model and its improvements, particularly in security and performance.

7/10

The video provides a detailed, hands-on review of Anthropic's new models, including pricing, benchmarks, and a practical comparison with GPT-5.6. The host acknowledges the limitations of self-reported benchmarks and supplements with independent data from Artificial Analysis. However, the review is subjective and based on a single, non-controlled test.

Markers derived by PSI from the transcript: the creator did not define chapters.

• 02:00Introduction to Fable 5.1 and Mito 5.1, explaining they are the same model with different safety restrictions.
• 05:00Discussion of the previous Fable 5 controversy and the government block.
• 08:00Overview of pricing changes, including 25% cost reduction and 75% cache read cost reduction.
• 10:00Presentation of benchmark results from Anthropic and independent Artificial Analysis.
• 12:00Explanation of Fable 5.1's ability to avoid shortcuts and solve root causes of software issues.
• 14:00Showcase of customer testimonials, including a rare bug discovery.
• 16:00Discussion of scientific research capabilities, including protein design and Venus mapping.
• 18:00Start of the practical test: redesigning the EdCamp 11 landing page with a vague prompt.
• 20:00Comparison of Fable 5.1 and GPT-5.6 outputs, highlighting design and copy differences.
• 22:00Final thoughts on the models' performance and the host's appreciation for Anthropic's new messaging.

• EDteam free courses — Promotional link for free courses, mentioned in the video description.
• EDteam courses — Link to all EDteam courses, mentioned in the video description.
• EDteam student scholarships — Link for student scholarships, mentioned in the video description.
• EDteam Instagram — Social media link, mentioned in the video description.
• EDteam LinkedIn — Social media link, mentioned in the video description.
• EDteam premium — Link to premium subscription, mentioned in the video description.
• EDteam teachers — Link for potential instructors, mentioned in the video description.
• EDteam TikTok — Social media link, mentioned in the video description.

• Artificial Analysis — Independent benchmark site referenced in the video, showing Fable 5.1 leading in intelligence.

• OpenAI GPT-5.6 — The host notes that GPT-5.6 performed comparably and even better in some aspects, such as design aesthetics, challenging the notion that Anthropic is overwhelmingly superior.