Anthropic has backtracked on plans to shift its most powerful AI model, Claude Fable 5, to a fully metered billing system, announcing late Friday that the model will remain a standard feature for subscribers on its top-tier plans. The decision marks the third policy reversal in two weeks and reflects the intense competitive pressure from OpenAI’s newly launched GPT-5.6 model.<\/p>\n
Starting July 20, Fable 5 will be included in the Max plan, priced at $100 per month, and the Team Premium plan, at $125 per user per month, with usage capped at 50% of each plan’s regular limits. Users on the lower-cost Pro plan at $20 per month and Team Standard plan at $25 per user per month will lose base access to Fable 5 and must purchase usage credits to continue using the model. Anthropic said it will provide affected users with a one-time $100 credit.<\/p>\n
\u201cDemand for Fable has been challenging to predict, which is why we’ve rolled it out to subscription plans in stages, extending access several times as we secured additional capacity,\u201d Anthropic said in a post on Claude’s official X account. \u201cWe know this has caused frustration. We want to give more certainty about what’s included in each plan.\u201d<\/p>\n
Anthropic originally planned to move Fable 5 to a pure usage-credit system on July 7, then pushed the deadline to July 12 and again to July 19. Each extension was announced just hours before the cutoff, creating confusion among developers who had been racing to complete projects during what became known as \u201cFable Time.\u201d<\/p>\n
The chaotic rollout comes against a backdrop of rapidly intensifying competition. OpenAI released GPT-5.6 on July 9, splitting the model into three distinct tiers\u2014Sol, Terra, and Luna\u2014with pricing that significantly undercuts Anthropic. Sol charges $5 per million input tokens and $30 for output, half of Fable 5’s $10 and $50. Luna, the cheapest tier at $1 per million input tokens, already outperforms Anthropic’s Claude Opus 4.8 on coding benchmarks.<\/p>\n
Independent testing has shown a tight race between the two flagship models. On the Artificial Analysis Coding Agent Index, GPT-5.6 Sol scored 80 against Fable 5’s 77.2, while on the broader Intelligence Index, Fable 5 edged ahead by a single point. Hands-on evaluations have found Fable 5 superior on creative writing and single-shot coding tasks, producing richer outputs with music, animation, and power-ups that Sol’s versions lacked.<\/p>\n
\u201cIf Fable exits subscriptions after July 19, Anthropic’s best model for paying subscribers becomes Opus 4.8\u2014which Luna already beats on coding at a fraction of the price,\u201d analysts noted. \u201cKeeping Fable available, even at 50% of weekly limits, is the only thing keeping Anthropic’s subscription tier from looking worse than OpenAI’s mid-range on paper.\u201d<\/p>\n
Fable 5 has endured a particularly rocky path since its June 9 debut. The U.S. government ordered it shut down on June 12 after Amazon researchers discovered a jailbreak vulnerability. Anthropic pulled the model globally for 19 days, built a new safety classifier, and restored service on July 1 with a compressed access window and tighter weekly caps.<\/p>\n
Industry observers see the pricing turmoil as evidence of a deeper infrastructure gap. \u201cAnthropic is reportedly pursuing deals to lease computing power from SpaceX and Meta,\u201d one industry source said. \u201cUltimately, competitiveness in AI will be decided by infrastructure and computing power.\u201d<\/p>\n
The Max plan’s continued inclusion of Fable 5 preserves a key selling point for Anthropic’s premium tier. At $100 per month, Max subscribers retain access to what many developers consider the best model for creative and complex coding work, even as OpenAI’s three-tier strategy offers a wider range of price points.<\/p>\n
Anthropic said it continues to invest in new capacity and will provide updates as circumstances change. For now, the company has chosen to protect its highest-value subscribers rather than push through a full migration to usage-based billing, a decision that keeps Fable 5 in the hands of its most dedicated users while leaving the broader market to weigh the growing price-performance gap with OpenAI’s newest offerings.<\/p>\n","protected":false},"excerpt":{"rendered":"Anthropic has backtracked on plans to shift its most powerful AI model, Claude Fable 5, to a fully…\n","protected":false},"author":2,"featured_media":111069,"comment_status":"","ping_status":"","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[8],"tags":[53,38051,34352,57127,157,57129,57128,57130],"class_list":["post-111068","post","type-post","status-publish","format-standard","has-post-thumbnail","category-anthropic","tag-anthropic","tag-claude-fable-5","tag-gpt-5-6","tag-max-plan","tag-openai","tag-pro-plan","tag-team-premium-plan","tag-team-standard-plan"],"_links":{"self":[{"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/posts\/111068","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/comments?post=111068"}],"version-history":[{"count":0,"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/posts\/111068\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/media\/111069"}],"wp:attachment":[{"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/media?parent=111068"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/categories?post=111068"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/tags?post=111068"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}