Moonshot AI's Kimi K3, released July 16, 2026, is the largest open-source model yet released and ranked at or near the top of independent benchmarks, approaching the performance of Anthropic's Fable 5 at a fraction of the cost: $15 per million output tokens versus $50 for Fable, while DeepSeek-V4-Pro prices at $0.87. Enterprise adoption is now documented: Coinbase halved its AI spending by shifting to Kimi and Z.ai GLM models; DoorDash delegates coding tasks to Kimi; Airbnb uses Alibaba Qwen for customer service; and Chinese models accounted for 57% of tokens used by US firms on OpenRouter in one week in July. Structurally, the combination of open-source licensing, compute-constrained efficiency engineering, and domestic chip compatibility has created a cost floor that US closed-source vendors cannot easily match, compressing addressable revenue per token for US frontier labs. Financial institutions with exposure to US AI infrastructure spend or frontier-model revenue should model a scenario where Chinese models capture a sustained and growing share of enterprise inference workloads. Also reported by: https://www.bloomberg.com/news/articles/2026-08-03/alibaba-drops-another-china-ai-model-with-breakthrough-performance · https://www.scmp.com/tech/article/3362738/alibabas-ai-model-qwen38-max-made-widely-accessible-ahead-open-weights-release
580 other tracked developments touching the same claims.