Company Encyclopedia
View More
name
DeepSeek
DPSK.NA
A large model company under High-Flyer Quantification and is committed to exploring the essence of AI.
News
View More

The second wave of DeepSeek impact: V3.2 rewrites China's cloud ecosystem and chip ecosystem

JP Morgan stated that DeepSeek V3.2 achieves architectural revolution through sparse attention mechanisms, with reasoning capabilities reaching GPT-5 levels. The cost is only $0.28 per million tokens input, and API prices have decreased by 30-70%, with long text reasoning costs plummeting by 6-10 times. More importantly, it has achieved "Day-0" support for domestic chips such as Huawei Ascend and Cambricon for the first time, breaking the dependence on Nvidia

Wallstreetcn·
SH
688256
-1.02%
NA
DPSK
0.00%
Wallstreetcn·
SH
688256
-1.02%
NA
DPSK
0.00%

The AI Gold Rush: Models Proliferate, Scaling Laws Shift, and Turf Wars Erupt

The AI landscape is rapidly evolving with new model releases, debates on scaling laws, and emerging turf wars. The podcast 'Mixture of Experts' explores these developments, highlighting models like Mistral 3 and DeepSeek-V3.2, and Amazon's decision to block ChatGPT's shopping agent. The discussion emphasizes specialization, strategic positioning, and the shift towards smarter AI approaches amid business incentives and potential legal battles.

StartupHub·
US
AI
-1.57%
NA
DPSK
0.00%
StartupHub·
US
AI
-1.57%
NA
DPSK
0.00%

Behind the global trillion-token interaction: Open-source models shift from "one dominant player" to "a multitude of contenders," with China’s AI entering the top tier

OpenRouter platform's trillion-token data shows that the open-source model market share has risen to 33%, breaking the closed-source monopoly. China's open-source AI has achieved a historic breakthrough, with market share soaring from 1.2% to nearly 30% peak, with Qwen, DeepSeek, and others entering the global first tier. The market landscape has shifted from "one dominant player" to a multi-strong co-governance, with medium models (15B-70B) becoming mainstream, and agent reasoning surpassing text generation, with programming applications accounting for over 50%

Wallstreetcn·
NA
DPSK
0.00%
Wallstreetcn·
NA
DPSK
0.00%

The Great AI Model Race Intensifies

The AI model race has intensified with major releases from Google, Anthropic, and DeepSeek, causing a competitive crisis at OpenAI. Google launched Gemini 3, Anthropic released Claude Opus 4.5, and DeepSeek unveiled V3.2. OpenAI's CEO issued a "Code Red" to improve ChatGPT. The competition is reshaping enterprise AI strategies, with lifecycle management and governance protocols becoming crucial.

Talkmarkets·
US
AI
-1.57%
NA
OpenAI
0.00%
NA
DPSK
0.00%
Talkmarkets·
US
AI
-1.57%
NA
OpenAI
0.00%
NA
DPSK
0.00%

Mistral Roars Back With Frontier AI Family That Goes Head to Head With DeepSeek

French startup Mistral releases a 4-model AI family, challenging DeepSeek with frontier performance and EU data sovereignty for enterprises wary of Chinese alternatives. source:

CoinLive·
NA
DPSK
0.00%
CoinLive·
NA
DPSK
0.00%
Posts
View More

The US has roughly 5,381 data centers compared to China's 449. That's more compute infrastructure than literally every other major country combined. The American thesis is straightforward whoever owns...

.....................
  • 0

Deepseek scare: market is up 10.4% since

Tariff scare: 23%

There are always reasons to sell.

Taiwan has warned citizens to beware of DeepSeek and other China generative AI language models due to misinformation and potential security threats, media report. The National Security Bureau inspecte...

Baidu 1Q25 Quick Interpretation: The first-quarter performance slightly exceeded expectations overall (cloud, user numbers). Simply put, DeepSeek was a big help!

1. The main highlight was the Smart Clo...

  • 0
© 2025 Longbridge｜Disclaimer