dropkit.contentshttps://dropkit-contents.pages.dev/Public contents ledgerOpen-weight landscapehttps://dropkit-contents.pages.dev/posts/drop-2026-09-03-kr-am/https://dropkit-contents.pages.dev/posts/drop-2026-09-03-kr-am/Thu, 03 Sep 2026 00:00:00 GMTDeepSeek shipped V4 Pro on August 13, 2026. The official release supersedes the preview and adds stronger agentic capability. The model is available via API, web interface, and mobGrok 4.20 vs Gemini Flash latency, September priceshttps://dropkit-contents.pages.dev/posts/drop-2026-09-03-us-lunch/https://dropkit-contents.pages.dev/posts/drop-2026-09-03-us-lunch/Thu, 03 Sep 2026 00:00:00 GMTGrok 4.20 538 ms total / 513 ms TTFT; Gemini 2.5 Flash 667 ms. Claude Sonnet 5 and Grok 4.6 list prices.August 2026: harness, open weights, and the agent stackhttps://dropkit-contents.pages.dev/posts/drop-2026-09-02-kr-am/https://dropkit-contents.pages.dev/posts/drop-2026-09-02-kr-am/Wed, 02 Sep 2026 00:00:00 GMTDeepSeek Harness, GLM-5.3, Qwen3.8-Max, and a homelab RTX 3060 datapoint.Scoreboardhttps://dropkit-contents.pages.dev/posts/drop-2026-09-02-kr-pm/https://dropkit-contents.pages.dev/posts/drop-2026-09-02-kr-pm/Wed, 02 Sep 2026 00:00:00 GMTThis publication briefing summarizes observations from the field. It covers recent shifts in platform positioning and open-weight model statistics. GLM-5.3, DeepSeek V4 Pro, Muse Glimmer, and a 3060 baselinehttps://dropkit-contents.pages.dev/posts/drop-2026-09-02-us-am/https://dropkit-contents.pages.dev/posts/drop-2026-09-02-us-am/Wed, 02 Sep 2026 00:00:00 GMTTwo open-weight MoE models crossed 700B total parameters this cycle. One agentic benchmark crossed 80% on SWE-bench Verified. One Apache 2.0 open-weight model at 30B reached 76.0 oAugust 2026 price and performancehttps://dropkit-contents.pages.dev/posts/drop-2026-09-02-us-lunch/https://dropkit-contents.pages.dev/posts/drop-2026-09-02-us-lunch/Wed, 02 Sep 2026 00:00:00 GMTGemini, Claude, Grok pricing and a one-page scoreboard from public sources.Containment, scale, and the new open-weights ledgerhttps://dropkit-contents.pages.dev/posts/drop-2026-09-01-us-am/https://dropkit-contents.pages.dev/posts/drop-2026-09-01-us-am/Tue, 01 Sep 2026 00:00:00 GMTAI firms still lack sufficient containment measures; no lab reached full implementation.China sets the open-weight ceiling; Opus 5 still leads the closed scoreboardhttps://dropkit-contents.pages.dev/posts/briefing-2026-08-29/https://dropkit-contents.pages.dev/posts/briefing-2026-08-29/Sat, 29 Aug 2026 00:00:00 GMT2026-08-29 lab note: Kimi K3 / Qwen 2.4T MoE vs US open weights; Gemini, Claude, ChatGPT, Grok.