NewTimeSpace | Hong Kong Market Close: The Three Major Indices Opened Lower and Moved in a Choppy Range, with the Hang Seng Index Closing Down 0.29%

NewTimeSpace reported, on September 24, 2026, the three major Hong Kong indices opened lower and moved in a choppy range, with the Hang Seng Index down 0.29%, the Hang Seng TECH Index down 0.41%, and the Hang Seng China Enterprises Index down 0.09%. The Hang Seng Index's market turnover today was 159.6 billion HKD, lower than that of the previous trading day. Southbound funds saw a net inflow of 2.900 billion HKD.

On the market, by industry. Themes such as oil and petrochemicals, paper & packaging, and public utilities strengthened against the trend. Themes such as electrical equipment, non-ferrous metals, and pharmaceuticals and biotechnology led the declines.

Among the Hang Seng Index constituents, 36 rose and 57 fell. On the downside, CMOC fell 3.70%, CATL fell 3.66%, and LONGFOR GROUP fell 3.62%. On the upside, PETROCHINA rose 2.83%; NEW ORIENTAL-S rose 2.63%, and CNOOC rose 2.14%.

Among the Hang Seng TECH Index constituents, 13 rose and 16 fell. On the downside, HUA HONG GRACE fell 3.33%, TRIP.COM-S fell 3.09%, and BIDU-W fell 2.54%. On the upside, SENSETIME-W rose 4.42% and HORIZONROBOT-W rose 2.22%. XIAOMI-W rose 1.53%.

Among Hong Kong Stock Connect constituents, DMALL surged 15.62%, TUHU-W surged 11.31%, and CREALIGHTS rose 9.42%. On the downside, HBM HOLDINGS-B fell 9.25%, METIS TECHBIO-P fell 8.98%, and INNOCARE fell 7.87%.

KAIYUAN SECURITIES pointed out that since August 2026, the global AI industry has entered a stage in which three lines run in parallel: domestic models intensively challenging the global first tier, API pricing shifting from one-sided decline to peak-valley tiering, and flagship price bands being re-rated. The successive breakthroughs of the GLM-5.3 series and GPT-Astra are not merely a leap in parameters and leaderboard numbers, but mark models' two-way breakout toward both Pareto optimality and frontier intelligence improvement, powerfully proving that the ceiling of open-source models is likewise very high, that the international competitiveness of domestic models has further improved, and that a balance between performance and cost is expected to open up the ceiling of global AI demand. The implementation of multimodality and the scaling of Agents are driving long-term rigid expansion of demand for training and inference computing power. (Source: "Hong Kong Stock Industry In-Depth Report: The Market Places Greater Emphasis on Model Pareto Optimality; Model Makers, Cloud Makers, and Sovereign Funds Have Ample Capital Ammunition" by KAIYUAN SECURITIES, 20260924)

NewTimeSpace Disclaimer: All content herein is the original work of NewTimeSpace. Any reproduction, reprinting, or use of this content in any other manner must clearly indicate the source as "NewTimeSpace". NewTimeSpace and its authorized third-party information providers strive to ensure the accuracy and reliability of the data, but do not guarantee the absolute correctness thereof. This content is for reference only and does not constitute any investment advice. All transaction risks shall be borne by the user.