Company
DeepSeek
DeepSeek is the Chinese lab whose low-cost, open-weight models rival far more expensive frontier systems.
DeepSeek prices web search for AI agents at RMB 5 per 1,000 calls
The Chinese lab opens its in-house search engine to all developers, charging only for successful requests and leaving some agent traffic free until November.
DeepSeek V4.1-Flash-1017 lifts benchmarks under an MIT licence
DeepSeek gives away a cheaper-to-run Flash update with a 1M-token context, though Opus 5 still leads on most listed tests.
DeepSeek's $12B round puts Tencent and CATL behind a 2027 IPO
Bloomberg reports the Chinese AI lab is finalising at least 80 billion yuan from strategic backers before a planned Hong Kong listing.
DeepSeek bets $2.5bn on Huawei to run its models at scale
A 160,000-chip Ascend order in Inner Mongolia shows China's AI leaders building inference capacity around Nvidia, not through it.
DeepSeek's benchmark chart is doing work its own harness can't back up
0813 is a genuinely good, genuinely cheap coding model that ranks last of seven on agentic coding. DeepSeek would rather you noticed the first half.
DeepSeek's V4-Flash runs 100x cheaper than Claude Fable 5
Artificial Analysis finds the Chinese model averages 3 cents per benchmark test against $3.15 for Anthropic's flagship, sharpening the price war over routine AI work.
An 80% cut three weeks after launch isn't generosity. It's a floor being set by DeepSeek
OpenAI says efficiency. The calendar says Claude Opus 5, Gemini 3.6 Flash and a Chinese model at fourteen cents.
Kimi K3 is 'open' the way a container ship is portable
2.8 trillion parameters, a custom licence, and a 64-accelerator minimum. The weights are public. The capability is not.
Chinese open models cut US firms' AI bills in half
Coinbase, DoorDash and Airbnb are routing routine AI work to Moonshot, Z.ai and DeepSeek, squeezing the margins Anthropic and OpenAI charge for commodity work.
Tokenmaxxing is over: Chinese open models are now 61% of top API traffic
The trillion-dollar labs sold you 'just spend, prices always fall.' In June it broke in public.
Open weights caught the frontier. The moat is leaking.
When you can download a near-top-tier model for free, 'exclusive access' stops being a business.
DeepSeek V4: the benchmark is the distraction. The chip pivot isn't.
1.6 trillion parameters, open weights, cheaper than the competition — and the real news is none of that.