Anthropic is facing a major problem. While Fable 5\u2014by far their most powerful AI model\u2014is supposed to be accessible only to U.S. users, the company behind Claude chose to withdraw the LLM entirely. Weeks later, Claude Sonnet 5 has now surfaced, a release that seems somewhat anticlimactic at first glance.<\/p>\n
Benchmarks from an AI model developer often set the tone. Independent tests later reveal just how good a new LLM really is. This also applies to Claude Sonnet 5, which has been available for a few hours as of this writing. Sonnet 5 and Opus 4.8\u2014a larger and more expensive model\u2014are virtually neck-and-neck on the benchmarks. Both perform roughly equally well in coding, reasoning, computing tasks, and knowledge work. Anthropic emphasizes the agentic nature of the latest model, which is better than ever.<\/p>\n
Don\u2019t worry about security<\/p>\n
For years, Anthropic has cultivated a habit of generating an ungodly amount of fear, uncertainty, and doubt surrounding new model releases. Nothing could be further from the truth this time: Sonnet 5 is actually meant to come across as harmless and mundane. The U.S. government banned the use of Fable 5 and Mythos 5 by non-U.S. citizens due to an alleged threat to national security. A \u201ccyber-superweapon\u201d\u2014that\u2019s how Anthropic CEO Dario Amodei described the characterization he heard from early users of Mythos Preview. After countless models hit the market without much scrutiny, that unchecked pace has been interrupted.<\/p>\n
With Sonnet 5, Anthropic apparently has little to fear. Even though there are additional security guardrails\u2014more so than with its predecessor, Sonnet 4.6\u2014these measures are merely on par with the protections Anthropic provided for the use of Opus 4.7 and 4.8.<\/p>\n
Returning to the restrictions: OpenAI is also feeling the consequences, with a limited rollout of GPT-5.6 in its three variants. GPT-5.6 Sol Ultra can be viewed as a kind of Fable (and scores higher than Mythos\/Fable on OpenAI\u2019s own benchmarks); Sol itself is a kind of Opus; Terra is a potential competitor to the Sonnet series; and Luna is a compact model similar to Gemini 3.5 Flash or Claude Haiku.<\/p>\n
Not quite Opus and not quite attractive enough?<\/p>\n
According to Anthropic, Sonnet 5 bridges the gap with the more powerful Opus 4.8\u2014but at a lower price. According to Anthropic, it\u2019s a significant improvement over its predecessor, Sonnet 4.6, in areas such as complex reasoning, tool usage, coding, and knowledge work. Starting today, it\u2019s the default model for Free and Pro users and available to Max, Team, and Enterprise customers, as well as in Claude Code and on the Claude Platform. With a pop-up focused on everyday AI use\u2014such as planning a vacation\u2014Anthropic also appears to be pursuing a Gemini- or ChatGPT-like approach alongside this release.<\/p>\n
The pricing\u2014which is more relevant to professional users who typically focus on coding\u2014is noteworthy. Through August 31, Anthropic is charging an introductory rate of $2 per million input tokens and $10 per million output tokens. After that, the rates will increase to $3 and $15, respectively. By comparison, Opus 4.8 costs $5 and $25 per million tokens. Between the two models, users can adjust the effort level to balance cost and performance.<\/p>\n
Of course, that doesn\u2019t tell the whole story. Consider GLM-5.2 from the Chinese company Z.ai, an open-source model that, based on figures from the renowned Artificial Analysis, performs on par with some recent Claude Opus models. However, it is five times cheaper per token\u2026 but consumes far more tokens per output than Opus typically does. In that regard, Sonnet 5 could prove to be an interesting competitor via the API.<\/p>\n
Safer in an agentic context<\/p>\n
Anthropic states that Sonnet 5 exhibits less undesirable behavior than Sonnet 4.6 and is more resistant to prompt injection attacks. The model also hallucinates less. Nevertheless, in the internal audit, it shows a higher percentage of \u201cmisaligned\u201d behavior than the more capable Opus 4.8 and Mythos Preview.<\/p>\n
When it comes to cybersecurity tasks, Sonnet 5 actually performs deliberately poorly. In a way, it\u2019s quite comical to see Anthropic emphasizing this topic: don\u2019t expect much from this brand-new model in this specific area. In a test developed in collaboration with Mozilla to write exploits for Firefox vulnerabilities, the model never succeeded in building a working exploit. Furthermore, Anthropic has enabled cybersecurity safeguards by default\u2014the same ones used in Opus 4.7 and 4.8.<\/p>\n
Of course, this caution didn\u2019t come out of nowhere. Fable 5 and Mythos 5 have been blocked since mid-June<\/a> and have shown no signs of returning. Meanwhile, the policy has shifted to controlled access<\/a> for Mythos 5 in particular, with the U.S. government determining which organizations are allowed to use the most powerful models.<\/p>\n Modified tokenizer<\/p>\n Sonnet 5\u2019s seemingly weaker performance on paper compared to Fable 5 and Opus 4.8 is to be expected given this smaller-scale LLM. Nevertheless, Sonnet models have built a solid reputation for helping users more easily stay within their five-hour and weekly token budgets under their subscription plans. Now, the step forward appears to be primarily architectural.<\/p>\n Sonnet 5 uses a new tokenizer that processes text differently. As a result, the same input can yield more tokens\u2014approximately 1.0 to 1.35 times as many, depending on the type of content. Anthropic states that the introductory price has been set so that the transition is roughly cost-neutral. Rate limits have also been increased in Chat, Cowork, Claude Code, and the Claude Platform. Developers can access the model via the Claude API as claude-sonnet-5.<\/p>\n","protected":false},"excerpt":{"rendered":"Anthropic is facing a major problem. While Fable 5\u2014by far their most powerful AI model\u2014is supposed to be…\n","protected":false},"author":2,"featured_media":91242,"comment_status":"","ping_status":"","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[8],"tags":[179,1276,53,3154,182,2798,48047,415,31869],"class_list":["post-91241","post","type-post","status-publish","format-standard","has-post-thumbnail","category-anthropic","tag-agentic-ai","tag-ai-models","tag-anthropic","tag-anthropic-claude","tag-claude","tag-claude-code","tag-claude-sonnet-5","tag-llm","tag-opus-4-8"],"_links":{"self":[{"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/posts\/91241","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=91241"}],"version-history":[{"count":0,"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/posts\/91241\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/media\/91242"}],"wp:attachment":[{"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/media?parent=91241"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/categories?post=91241"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/tags?post=91241"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}