$ZHIPU listed and surged from 167 to 149 in just three days. To be honest, such volatility isn’t that outrageous in Hong Kong stock warrants, but for tokenized assets, many people’s nerves have already cracked.
I think we need to break this down into two layers. The first layer is Zhipu’s own fundamental base. The GLM model has the most solid progress in commercial deployment in China—it has customers, and it has scenarios. The second layer is market sentiment. The AI narrative has been cold on the US stock market for over half a year; $NVDA has been digesting valuations through choppy consolidation. When domestic large-model companies list at this moment, they’re mainly absorbing liquidity overflow rather than attracting incremental hot money.
The 153 level is quite tricky. Support below at 149.74 is the IPO low—if it breaks, it’s a true breakdown, and downside room opens toward around 140. Resistance above at 167.9 is a short-term trapped-holder dense zone. Any rebound that can’t hold and stay above 165 is considered weak.
My bias is bearish, but I won’t short. In products with shallow liquidity, chasing a short can easily get “squeezed” by a counter-push. If you want to trade it, wait for a pullback into the 149–150 area to see whether there’s decent support and follow-through. If it shows shrinking volume and stabilizes, you can try a small long position. Set a stop-loss at 146.5. If it breaks 149 directly on expanding volume, don’t bottom-fish—wait and reassess around 140–142.
The trading value of 25M suggests the chips haven’t fully rotated yet. At this stage, the volatility isn’t suitable for heavy positioning. In plain terms, tokenized assets give you 7x24 trading convenience, but they don’t give you deeper liquidity—so weigh your position size accordingly.
I think we need to break this down into two layers. The first layer is Zhipu’s own fundamental base. The GLM model has the most solid progress in commercial deployment in China—it has customers, and it has scenarios. The second layer is market sentiment. The AI narrative has been cold on the US stock market for over half a year; $NVDA has been digesting valuations through choppy consolidation. When domestic large-model companies list at this moment, they’re mainly absorbing liquidity overflow rather than attracting incremental hot money.
The 153 level is quite tricky. Support below at 149.74 is the IPO low—if it breaks, it’s a true breakdown, and downside room opens toward around 140. Resistance above at 167.9 is a short-term trapped-holder dense zone. Any rebound that can’t hold and stay above 165 is considered weak.
My bias is bearish, but I won’t short. In products with shallow liquidity, chasing a short can easily get “squeezed” by a counter-push. If you want to trade it, wait for a pullback into the 149–150 area to see whether there’s decent support and follow-through. If it shows shrinking volume and stabilizes, you can try a small long position. Set a stop-loss at 146.5. If it breaks 149 directly on expanding volume, don’t bottom-fish—wait and reassess around 140–142.
The trading value of 25M suggests the chips haven’t fully rotated yet. At this stage, the volatility isn’t suitable for heavy positioning. In plain terms, tokenized assets give you 7x24 trading convenience, but they don’t give you deeper liquidity—so weigh your position size accordingly.