awaited.dev ยท Hands-on notes on AI and software developmentModel releases, benchmarks, experiments and guides for developers who work with AI every day. Written from real use, with sources for every claim.https://awaited.dev/enSat, 01 Aug 2026 00:00:00 GMTA.M.Astrohttps://www.rssboard.org/rss-specificationCoding agents start every session blank. Mine get a maphttps://awaited.dev/guides/context-management-for-coding-agents/https://awaited.dev/guides/context-management-for-coding-agents/AI agents forget my repo between sessions, so I built a four-layer file map that points them to current facts. Here is how it works and what it costs.Thu, 30 Jul 2026 00:00:00 GMTA.M.guidescontext-managementcoding-agentsclaude-codeagents-mddocumentationvibe-codingKimi K3 matches GPT-5.6, but takes 4.6 times longerhttps://awaited.dev/benchmarks/kimi-k3-glm-5-2-hype-check/https://awaited.dev/benchmarks/kimi-k3-glm-5-2-hype-check/Kimi K3 matches GPT-5.6 on the Coding Agent Index, but needs 4.6 times longer to do it. I checked whether either Chinese model is truly cheaper.Thu, 30 Jul 2026 00:00:00 GMTA.M.benchmarkskimi-k3glm-5-2benchmarkscoding-agentsopen-weightspricingOpus 5 ties Sol, Terra matches Fable at a quarter the pricehttps://awaited.dev/benchmarks/gpt-5-6-vs-claude-5-benchmarks/https://awaited.dev/benchmarks/gpt-5-6-vs-claude-5-benchmarks/I compare Sol, Terra, Opus 5 and Fable 5 across coding benchmarks to show why the top score, lowest cost and right effort setting rarely align.Thu, 30 Jul 2026 00:00:00 GMTA.M.benchmarksgpt-5-6opus-5fable-5benchmarkscoding-agentsreasoning-effortI compared Convex with eight alternatives. Convex stayshttps://awaited.dev/guides/convex-alternatives/https://awaited.dev/guides/convex-alternatives/Most of what I build with AI runs on Convex. I compared it with eight alternatives, checked their prices, and found one costly issue with EU hosting.Wed, 29 Jul 2026 00:00:00 GMTA.M.guidesconvexsupabaseappwriteinsforgebackendvibe-codingOpus 5 is a great employee and a frustrating bosshttps://awaited.dev/releases/opus-5-great-employee-bad-boss/https://awaited.dev/releases/opus-5-great-employee-bad-boss/Opus 5 matches Fable 5 on benchmarks at half the token price. I looked at why users still find it frustrating, and when it needs a stronger model.Wed, 29 Jul 2026 00:00:00 GMTA.M.releasesopus-5fable-5anthropicbenchmarksclaude-codeGPT-5.6 Sol finds everything. That's the problemhttps://awaited.dev/experiments/gpt-5-6-sol-overengineering/https://awaited.dev/experiments/gpt-5-6-sol-overengineering/GPT-5.6 Sol found nearly six times as many audit issues as Fable 5, but most failed triage. Here is how I use its persistence without losing control.Wed, 29 Jul 2026 00:00:00 GMTA.M.experimentsgpt-5-6solopenaicodexfable-5multi-agent