Claude Fable 5.1 Scores Tops Artificial Analysis Intelligence Index With Score Of 66, Beats Opus 5 By 3 Points

Anthropic's Claude Fable 5.1 topped the Artificial Analysis Intelligence Index with a score of 66, beating its predecessor Claude Opus 5 by 3 points and leading the industry. OpenAI's GPT-5.6 Sol and xAI's Grok 4.6 tied for third at 61, while Google's Gemini 3.7 Flash scored 56. The result follows Anthropic's rollout of Fable 5.1 and Mythos 5.1, which included a 75% cut to cache read pricing and new safety features.

Fable 5.1 https://officechai.com/ai/fable-5-1-benchmarks/ has made a big jump and grown Anthropic’s lead in the Artificial Analysis Intelligence Index. According to the latest results published by Artificial Analysis, Claude Fable 5.1 running at max effort, with fallback has topped the benchmark’s Intelligence Index with a score of 66, putting it ahead of every other model currently tracked on the leaderboard. The next closest model is Anthropic’s own Claude Opus 5 https://officechai.com/wp-content/uploads/2026/07/HOAffmCWsAA1D X.png , which scored 63 — meaning Fable 5.1 beats Anthropic’s previous flagship by 3 points, and now holds a comfortable lead over the rest of the field. Where It Stands Against The Competition The gap to the rest of the industry is more pronounced. OpenAI’s GPT-5.6 Sol https://officechai.com/ai/openai-launches-gpt-5-6-sol-beats-mythos-on-terminalbench/ and xAI’s Grok 4.6 https://officechai.com/ai/grok-4-6-benchmarks/ high are tied for third place with a score of 61 apiece, five points behind Fable 5.1. Moonshot AI’s Kimi K3 max and Zhipu’s GLM-5.1 max follow closely with scores of 60 each, while MiniMax’s Spark 1.2 high trails slightly further behind at 57. Google’s Gemini 3.7 Flash https://officechai.com/ai/gemini-3-7-flash-is-at-pareto-frontier-of-intelligence-vs-speed-says-artificial-analysis/ high posted a score of 56, ahead of DeepSeek’s V4 Pro 0813 max at 53 and OpenAI’s GPT-5.6 Luna max at 52. NVIDIA’s Nemotron 3 Ultra rounds out the chart with a score of 38, well behind the rest of the pack. The Artificial Analysis Intelligence Index aggregates performance across a range of reasoning, coding, and knowledge benchmarks to produce a single comparative score, and is widely used as a quick reference point for how frontier models stack up against each other. Part Of A Broader Push The result comes just after Anthropic’s official rollout of Claude Fable 5.1 and Claude Mythos 5.1, which the company positioned as its most advanced models yet for coding, knowledge work, and early-stage scientific research. Alongside capability gains, that release brought a 75% cut to cache read pricing translating into roughly 25–45% lower overall costs depending on workload , a new enterprise data-retention framework called Enterprise Frontier Safeguards, and refined safety classifiers aimed at cutting down false positives in cybersecurity and biology-related queries. Anthropic had already published its own internal benchmark results showing Fable 5.1 ahead of Fable 5, Opus 5, and GPT-5.6 Sol across metrics like Terminal-Bench-Science 0.1, GDPval-AA v2, and Humanity’s Last Exam. The Artificial Analysis Intelligence Index score adds a third-party data point reinforcing that lead, with Fable 5.1 now holding the top spot on one of the industry’s most closely watched independent leaderboards.