OneStone AI BlogBuild-in-public engineering notes from OneStone — a local-first AI agent desktop for Linux and Windows.https://onestoneai.com/Frontier AI Prices Rose 100% Since January. Open Models Took 15% of the Market. The Builder Decision.https://onestoneai.com/blog/ai-inference-cost-split-2026/https://onestoneai.com/blog/ai-inference-cost-split-2026/Frontier token prices doubled since January 2026. Open-weight models captured 15% of inference in 12 months. The routing decision between the two tiers is where 60–80% of AI cost savings live.Mon, 10 Aug 2026 00:00:00 GMTThe Frontier Model Availability Gap: 63.6% of 2026 AI Launches Arrived Gatedhttps://onestoneai.com/blog/frontier-model-availability-gap-august-2026/https://onestoneai.com/blog/frontier-model-availability-gap-august-2026/Axis Intelligence Research tracked 11 frontier model launches in 2026 — 7 shipped behind a gate. Here is what the August frontier actually looks like for builders.Fri, 07 Aug 2026 00:00:00 GMTThe Benchmarks You Trusted to Pick Your AI Model Are Brokenhttps://onestoneai.com/blog/ai-benchmarks-are-broken-2026/https://onestoneai.com/blog/ai-benchmarks-are-broken-2026/MMLU is saturated. HumanEval is contaminated. SWE-bench swings 25 points based on scaffolding. Here is what actually signals model capability in 2026.Wed, 05 Aug 2026 00:00:00 GMTThe EU AI Act Landed Yesterday. Here's What Actually Changed for Builders.https://onestoneai.com/blog/eu-ai-act-august-2026-what-landed/https://onestoneai.com/blog/eu-ai-act-august-2026-what-landed/The high-risk AI deadline got pushed to December 2027. But Article 50 transparency rules went live August 2 — and most builders missed the deadline that actually landed.Mon, 03 Aug 2026 00:00:00 GMTLongCat-2.0 Beats GPT-5.5 on SWE-bench Pro — What It Meanshttps://onestoneai.com/blog/longcat-2-0-swebench-pro/https://onestoneai.com/blog/longcat-2-0-swebench-pro/A Chinese open-source model just outscored OpenAI's latest on coding. Here's why it matters for the competition.Tue, 07 Jul 2026 00:00:00 GMTAnthropic suspends Fable 5 and Mythos 5 over US export controlshttps://onestoneai.com/blog/anthropic-fable-mythos-export-suspension/https://onestoneai.com/blog/anthropic-fable-mythos-export-suspension/How a government compliance order cascades through the entire AI stack—and what it means for builders outside the US.Mon, 06 Jul 2026 00:00:00 GMTThe Line Between Free and Restricted AI Is Now a Benchmark Scorehttps://onestoneai.com/blog/ai-regulation-is-a-benchmark-score/https://onestoneai.com/blog/ai-regulation-is-a-benchmark-score/The White House is drafting voluntary release standards for frontier AI. The trigger is not a law — it is a cybersecurity eval score between 70.7% and 96.7%.Sat, 04 Jul 2026 00:00:00 GMT