{"slug": "anthropic-s-claude-fable-5-2-and-mythos-5-0-are-impressive-but-be-prepared-to-a", "title": "Anthropic's Claude Fable 5.2 and Mythos 5.0 Are Impressive, But Be Prepared to Sell a Kidney to Afford Them", "summary": "Anthropic released Claude Fable 5.2 and restricted sibling Mythos 5.2 on Monday, claiming output up to 30% faster than predecessors and a jump to 49.1% on Anthropic's internal FrontierCode 2.0 coding evaluation from 32.4% for Fable 5.1. AI YouTuber Matthew Berman's field tests estimated API costs of $14,710 for a Gears of War-style level (43 hours 48 minutes, 8.63 billion tokens), $11,700 for a browser-based Excel clone and $13,000 for a Fall Guys-style game, while Anthropic also said Fable helped solve two open theoretical-physics problems involving black holes and design three functioning electronic devices. Fable and Mythos share the same underlying model, but Fable carries stricter safeguards on dangerous biological and cybersecurity requests and Mythos is reserved for verified organizations in those fields.", "body_md": "#### In brief\n\n- Anthropic released Claude Fable 5.2 and restricted sibling Mythos 5.2, promising stronger coding and reasoning at lower per-token prices.\n- Matthew Berman's tests estimated $14,710 for a Gears of War-style level, $11,700 for an Excel clone, and $13,000 for Fall Guys.\n- Eligible paid Claude subscribers get temporary Fable access or usage resets, but heavy workloads can exhaust allowances quickly.\n\nAnthropic [released](https://claude.ai/chat/8e7f0ddd-db09-40a8-b37b-e6b29ec3e3df) Claude Fable 5.2 and Mythos 5.2 on Monday, promising AI models that can turn a handful of instructions into working software, design physical objects, and tackle research problems that previously required teams of specialists.\n\nYou can now ask a chatbot to build the app you've always wanted. You may also need to sell a kidney to pay for it.\n\nThat's not because every prompt costs thousands of dollars. It's because Fable can keep working for days, writing code, testing it, and fixing its own mistakes while the meter keeps running.\n\nSome users have already tested its capabilities, and they’ve been impressed—by the power, and the price.\n\nAI YouTuber Matthew Berman put the model through a series of ambitious tests and described Fable 5.2 as the “[best AI model](https://forwardfuture.com/newsletter/originals/i-put-fable-5-2-to-work-5-real-world-field-tests-and-what-i-learned)” he's ever used. His projects included playable video games and a browser-based version of Microsoft Excel, complete with formulas and pivot tables.\n\nOne single level of Gears of War: E-Day-inspired level took 43 hours and 48 minutes of build time and processed 8.63 billion tokens—small chunks of information an AI reads and produces. The running API cost estimate, based on pay-as-you-go pricing for developers, had reached $14,710.\n\nFable 5.2 is HERE! 🔥\n\nI've burned over 100 billion tokens early testing Fable 5.2. It is BY FAR the best model I've ever used.\n\nHere's my full review (including Gears of War clone):\n\n> Fable 5.2 is the most autonomous model to date. Given a task with little direction, it makes… [pic.twitter.com/0xTzqTY040](https://t.co/0xTzqTY040)\n\n— Matthew Berman (@MatthewBerman) [November 9, 2026](https://x.com/MatthewBerman/status/2119871207206158459?ref_src=twsrc%5Etfw)\n\nThe Excel clone ran up an estimated $11,700 after discounts for cached inputs, or information the system can reuse instead of processing from scratch. Berman also showed a Fall Guys-style game that consumed more than 9 billion tokens over 71 hours, equivalent to roughly $13,000.\n\nThose are estimated API costs, and the projects weren't flawless: Berman reported awkward animations, bugs, and features that still needed testing.\n\n## A $37 tower and a $15,000 video game\n\nIf you’re poor, well… Fable also knows how to work for cheap. For less than $40 (two months of Claude Pro) it built a cute orange cylinder.\n\nAnother experiment simulated water-balloon physics for $91. It’s worth noting this isn’t a model intended for routine tasks. This is for serious work and people with deep pockets.\n\nIn its [launch announcement](https://www.anthropic.com/claude-fable-and-mythos-5-2), Anthropic said the new models generate output up to 30% faster than their predecessors. The company also says Fable helped solve two open theoretical-physics problems involving black holes and design three functioning electronic devices, claims that will need independent scrutiny.\n\nAnthropic's internal FrontierCode 2.0 evaluation, which measures the percentage of proposed coding changes that would be accepted into real open-source projects, put Fable 5.2 at 49.1%, compared with 32.4% for Fable 5.1. That helps explain why users are pushing it beyond simple chat responses.\n\nFable and Mythos share the same underlying model, but Fable has stricter safeguards around dangerous biological and cybersecurity requests. Mythos is reserved for verified organizations working in those fields.\n\nThe standard API price is $8 per million input tokens and $40 per million output tokens, down from the previous generation's $10 and $50. Optional fast mode can generate output up to 2.5 times faster, but doubles the rates to $16 and $80.\n\nFor contrast, Kimi 3.1 Pro, a model considered by many as the best Chinese AI model costs $2 per million input tokens and $10 per million output tokens. The latest Deepseek model costs $0.6 and $1.98 outside of peak hours.\n\nPer Anthropic’s standards, Fable’s price is still twice the per-token price of Claude Opus 5.5, which Anthropic recommends for everyday work. Its flagship is meant for problems the cheaper models struggle to finish, not for asking whether your email sounds too aggressive.\n\nThe cost debate has history. When Fable 5 was [released](https://decrypt.co/370688/internet-furious-anthropic-claude-mythos-fable-5), it burned through some subscribers' quotas in minutes. When [Fable 5.1 arrived](https://decrypt.co/377078/anthropic-claude-fable-5) in September, Anthropic kept premium pricing while improving discounts for reused inputs.\n\nJust last week, [speculation swirled](https://decrypt.co/382390/claude-mythos-5-5-fable-5-5-gpt-6-2-astra-leaks) around a supposed Fable 5.5. The actual release is 5.2, with cheaper token rates but enough appetite to turn a hobby project into a serious expense.\n\nFor ordinary subscribers, [Anthropic's launch offer](https://support.claude.com/en/articles/17506494-claude-fable-5-2-launch-offers) gives Pro users and standard Team seats Fable access for one week, capped at half their weekly allowance, plus a reset. Max, Ultra and eligible premium seats get one to three resets depending on subscription history.\n\nPro users' promotional access ends November 16, after which Fable returns to paid credits on their plans. Resets for eligible Max and Ultra subscribers last until March 8, 2027. Berman's $14,710 game offers a fairly convincing reason to check the meter.\n\nAnthropic, last we checked, doesn’t accept kidneys as a method of payment.\n\n## Daily Debrief Newsletter\n\nStart every day with the top news stories right now, plus original features, a podcast, videos and more.\n\nJoin 72,000+ readers\n\n[Add on Google](https://www.google.com/preferences/source?q=decrypt.co)", "url": "https://wpnews.pro/news/anthropic-s-claude-fable-5-2-and-mythos-5-0-are-impressive-but-be-prepared-to-a", "canonical_source": "https://decrypt.co/382897/anthropic-claude-fable-5-2-and-mythos-5-2-impressive-and-expensive", "published_at": "2026-11-09 22:36:02+00:00", "updated_at": "2026-11-09 22:49:29.604291+00:00", "lang": "en", "topics": ["artificial-intelligence", "large-language-models", "ai-products", "generative-ai", "ai-agents"], "entities": ["Anthropic", "Claude Fable 5.2", "Mythos 5.2", "Matthew Berman", "FrontierCode 2.0", "Claude Pro"], "also_reported_by": [], "alternates": {"html": "https://wpnews.pro/news/anthropic-s-claude-fable-5-2-and-mythos-5-0-are-impressive-but-be-prepared-to-a", "markdown": "https://wpnews.pro/news/anthropic-s-claude-fable-5-2-and-mythos-5-0-are-impressive-but-be-prepared-to-a.md", "text": "https://wpnews.pro/news/anthropic-s-claude-fable-5-2-and-mythos-5-0-are-impressive-but-be-prepared-to-a.txt", "jsonld": "https://wpnews.pro/news/anthropic-s-claude-fable-5-2-and-mythos-5-0-are-impressive-but-be-prepared-to-a.jsonld"}}