IPO$1.5B raise
DeepSeek in Talks to Raise $1.5B, Then IPO
DeepSeek is reportedly in talks to raise $1.5 billion in private funding as a step toward an eventual IPO, following Goldman Sachs naming it a preferred Chinese AI model.
Jul 14, 2026Coverage Timeline
newest first- Jul 14, 2026IPODeepSeek in Talks to Raise $1.5B, Then IPO$1.5B raise→
- Jul 12, 2026AIGoldman Names Its Preferred Chinese AI ModelsPT HK$1,880 (Zhipu)→
- Jul 7, 2026AIChina's Open-Weight Models Keep Closing the Gap$98B China chip self-reliance push→
- Jun 29, 2026AIDeepSeek and Peking University Open-Source DSpark, Speeding LLM Inference by Up to 85%Up to 85% faster→
- Jun 26, 2026AIDeepSeek Open-Sources DSpark, Claiming 60-85% Faster Generation→
- Jun 20, 2026AIHuawei-Led Team Post-Trains DeepSeek's 1.6-Trillion-Parameter Model on Ascend Chips, Not Nvidia→
- Jun 16, 2026FUNDINGDeepSeek Raises $7.4B -- Becomes China's Most Valuable AI Startup at $50B+$7.4B→
- Jun 16, 2026FUNDINGDeepSeek's $7.4B Round Backed by Tencent, CATL and JD$7.4B→
All Coverage
AIPT HK$1,880 (Zhipu)
Goldman Names Its Preferred Chinese AI Models
Goldman Sachs research named Zhipu, DeepSeek and ByteDance its preferred Chinese AI models, ranked ahead of Alibaba and Tencent on speed-to-market, and recommended rotating out of Korean AI stocks into a long China AI basket.
Jul 12, 2026AI$98B China chip self-reliance push
China's Open-Weight Models Keep Closing the Gap
Tencent, Zhipu, Alibaba and DeepSeek are locked in an intensifying open-weight model race, backed by a domestic chip industry drawing roughly $98 billion in spending as China reduces reliance on Nvidia.
Jul 7, 2026AIUp to 85% faster
DeepSeek and Peking University Open-Source DSpark, Speeding LLM Inference by Up to 85%
DeepSeek and Peking University open-sourced DSpark, a speculative-decoding framework that boosts per-user LLM generation speed by 60-85% -- and up to 661% throughput under tight latency constraints -- without hardware upgrades or model retraining. Released under the MIT license and already live in DeepSeek's V4-Flash and V4-Pro production models, it is another Chinese efficiency breakthrough aimed at slashing the cost of serving AI.
Jun 29, 2026AI
DeepSeek Open-Sources DSpark, Claiming 60-85% Faster Generation
DeepSeek released DSpark, a 'semi-parallel' speculative-decoding module for its DeepSeek-V4 Flash and Pro models, and open-sourced DeepSpec, a full codebase for training and evaluating speculative-decoding draft models. The company claims generation runs 60-85% faster on Flash and 57-78% faster on Pro at the same throughput -- a fresh reminder that the open-weight challengers keep narrowing the efficiency gap.
Jun 26, 2026AI
Huawei-Led Team Post-Trains DeepSeek's 1.6-Trillion-Parameter Model on Ascend Chips, Not Nvidia
A research group including Huawei completed full-parameter post-training of DeepSeek's 1.6-trillion-parameter V4-Pro model using a cluster of at least 1,000 Huawei Ascend 910C chips -- demonstrating that a frontier-scale Chinese model can be trained end-to-end on domestic silicon. It's a milestone in China's drive to break free of US chips and the CUDA ecosystem.
Jun 20, 2026FUNDING$7.4B
DeepSeek Raises $7.4B -- Becomes China's Most Valuable AI Startup at $50B+
DeepSeek closed a $7.4B round backed by Tencent, CATL, JD.com, NetEase and IDG Capital, vaulting past a $50B valuation to become China's most valuable AI startup. The raise is a direct answer to the US frontier labs and a statement that the model race is now a two-superpower contest.
Jun 16, 2026FUNDING$7.4B
DeepSeek's $7.4B Round Backed by Tencent, CATL and JD
The largest AI financing out of China this year: $7.4B from a strategic syndicate that hands DeepSeek both capital and distribution. At a $50B+ valuation, it is now the anchor of China's frontier-model ambitions.
Jun 16, 2026