Tag · 22 stories
Every story tagged DeepSeek on AI Chat Daily.
DeepSeek is the Chinese AI lab whose V3 and R1 models reset expectations about how much compute a frontier model actually needs. We follow every DeepSeek release, the US-China AI competition implications, and the open-weights strategy.
Alibaba, Moonshot AI, and DeepSeek tied to five campaigns harvesting Claude's reasoning traces, with one Moonshot request routed from the Chinese military.
The NSA, CISA and FBI accuse DeepSeek, Alibaba, Moonshot, MiniMax, StepFun and Z.AI of siphoning capabilities from Claude, GPT, Gemini and Grok.
The Hangzhou model lab is preparing a mainland listing, choosing China's largest brokerage as sponsor rather than a Hong Kong or offshore route.
DeepSeek, ByteDance and Alibaba are among the Chinese firms racing to build in the Inner Mongolian city, dwarfing OpenAI's 10GW Stargate footprint.
Claude's maker crossed $11.5B in a single quarter and posted positive adjusted operating income as it lines up a fall listing.
US token prices have dropped nearly a quarter since mid-July as DoorDash and Airbnb shift workloads to Chinese models from Moonshot and DeepSeek.
Liang Wenfeng told Yicai that DeepSeek will keep releasing its frontier models as open weights, prioritising AGI over commercial returns.
GLM-5.2 and DeepSeek V4-Pro closed the cyber capability gap from 6-10 months to 4-7 months. Kimi K3 is next in line.
OpenAI's o3 scored 1.83 on a segregation scale where 2 is total sorting — and reasoning models were the worst offenders.
The Chinese lab's next open-weight model will run 2–3 trillion parameters, the largest ever released from China.
Chinese open models took 41% of Hugging Face downloads this spring, and the top six models on OpenRouter are all open — Claude Opus 4.7 sits seventh.
DeepSeek now handles a third of tokens on Vercel's gateway, yet Anthropic still captures more than half of AI spend on the platform.
The Chinese LLM developer has spent about a year on a silicon project targeting data center inference, per Reuters.
Artificial Analysis launches the first agentic AI benchmark, and the GB300 NVL72 takes the top spot against Hopper on DeepSeek V4 Pro.
The startup's script-scoring tool rated flop Christy above Sinners, which grossed $370 million and won an Oscar.
The Chinese AI lab behind R1 is moving from self-funded curiosity to venture-backed frontier player with a single round larger than most of its rivals' lifetime totals.
The AI gateway startup more than doubled its value in a year as token volume surged 5x.
A matrix multiply that takes three lines in PyTorch takes 50-plus in CUDA — and that gap is why AMD and Intel keep losing.
The Beijing lab's valuation has doubled to $20B in months as its Kimi K2.6 model becomes the second-most-used LLM on OpenRouter.
Liang Wenfeng's previously self-funded lab is raising to give employees equity, with the China Integrated Circuit fund leading.
Ten Chinese chip designers including Cambricon and Biren backed the model on day one as the sanctioned firm bets on open source to catch up.
The open-source flagship matches GPT-5.4 and Claude-Opus-4.6 on benchmarks while pricing V4-Pro at $1.74 per million input tokens.
The tightly edited AI news email engineers, founders, and investors actually open. One email. Every weekday. Five minutes to finish.
The briefing read inside teams at
