dropkit.contents

Ledger on GitHub. Hub here. Branches elsewhere.

← Drops · 2026-09-13

drop-2026-09-13-kr-am

claimlabelordernumbersURL
Kimi K3 hit #1 on Arena.ai’s Frontend Code leaderboard, ahead of Claude Fable 5 and GPT-5.6 Sol.🟢 robustrobust1679 vs 1631 vs 1618https://kimik2ai.com/k3/
Kimi K3 is described as a 2.8T-parameter open-weight model.🟢 robustrobust2.8Thttps://kimik2ai.com/k3/
DeepSeek V4-Pro was described as a 1.7T-class MoE with open weights and a low per-token price.⚠️ sensitivepartial1.7T; 0.435/0.87 per million tokenshttps://mungomash.com/ai/models/
Alibaba Qwen3.8-Max was reported as a 2.4T frontier model with a 1M-token context window.🟢 robustrobust2.4T; 1M tokenshttps://www.forbes.com/sites/tylerroush/2026/08/03/alibaba-unveils-qwen38-max-model-chinas-latest-ai-challenger-to-openai-and-anthropic/
OpenAI’s InferenceX comparison reportedly showed better AI work per watt than Nvidia Blackwell systems.⚠️ sensitivepartial1.5–1.9×https://chinaaibench.com/news/2026-08-26/
In the same comparison, DeepSeek R1 and Kimi K2.5 were reported at much higher per-user throughput than Blackwell systems.⚠️ sensitivepartial700 vs 169; 694 vs 182 tokens/shttps://chinaaibench.com/news/2026-08-26/
Chinese open-weight models were said to be closing the gap with top closed models on cyber/bio capability.⚠️ sensitivehypeno stable numberhttps://techcrunch.com/2026/08/04/open-weight-ai-models-are-catching-up-to-the-frontier-the-safety-gap-remains/
Kimi K3 was reported to rank fourth on the Artificial Analysis Intelligence Index.⚠️ sensitivepartial4thhttps://www.forbes.com/sites/drewbernstein/2026/08/03/chinese-ai-models-at-the-frontier/
DeepSeek V4-Flash was described as a faster, cheaper tier with an AA Intelligence Index near GPT-5.6 Luna.⚠️ sensitivehype50 vs 51; ~60% lower task costhttps://stochasticsandbox.com/posts/llm-encyclopedia-2026-08-08/
Kimi K3 was released in late July 2026 as a 2.8T-parameter open-weight frontier model and hit the Arena.ai Frontend Code leaderboard.🟢 robustrobust2.8T; 104B active; 1679 vs 1631https://kimik2ai.com/k3/
DeepSeek V4-Flash-0731 shipped under MIT/open-weight licensing with low API pricing and a 1M-token context window.🟢 robustrobust304B total; 13B active; ~$0.14/$0.28 per million tokens; 1M contexthttps://www.together.ai/models/deepseek-v4-flash-0731
Qwen3.8-Max / Qwen3.8-2.4T-A95B was reported as a 2.4T frontier model with 95B active parameters and 1M-token context.🟢 robustrobust2.4T total; 95B active; 1M contexthttps://www.forbes.com/sites/tylerroush/2026/08/03/alibaba-unveils-qwen38-max-model-chinas-latest-ai-challenger-to-openai-and-anthropic/
GLM-5.3-Flash was reported as a 321B/18B MoE open-weight model under MIT with strong coding benchmark scores.⚠️ sensitivepartial321B total; 18B active; 84.3 Terminal Bench 2.1; 63.4 DeepSWE 1.1https://forkast.news/chinese-open-weight-frontier-compresses-five-labs-thirty-days-two-licensing-models/
Chinese open-weight frontier releases compressed into roughly a 30-day window across four flagship models, with two MIT and two custom-revenue-gated licenses.🟢 robustrobust30 days; 4 models; 2 MIT; 2 customhttps://forkast.news/chinese-open-weight-frontier-compresses-five-labs-thirty-days-two-licensing-models/
Tencent’s Hy4-preview was a 770B-parameter MoE model with a 1M-token context window and Apache 2.0 release.