Tag

Deepseek

All articles tagged with #deepseek

technology2 days ago•89 min saved

DeepSeek slashes V4-Pro price by 75% amid Huawei chip supply dynamics

Chinese AI startup DeepSeek will permanently cut its flagship V4-Pro AI model price by 75%, keeping costs at a quarter of the original and API pricing at 0.025–6 yuan per million tokens. DeepSeek did not specify whether the cut is tied to a higher Huawei Ascend 950 chip supply, but the move comes as Huawei’s AI chips and U.S. export controls shape China’s high-end compute access; when V4 was launched, Pro pricing was up to 12x Flash owing to limited compute, with expectations of steeper declines once Ascend 950 supernodes are deployed more broadly later this year.

via Reuters|

#ai #chip-supply #deepseek

technology29 days ago•75 min saved

The AI price split: why the middle market is disappearing

OpenAI’s GPT-5.5 pricing (about $5 input, $30 output per million tokens) and DeepSeek’s open-weight V4-Pro/Flash offerings have created a two-cluster AI market, widening the cost gap between a closed, vendor-managed product and cheaper, open infrastructure. The once-smooth price-performance curve now shows a thinning middle, forcing developers to route workloads across the high-end integrated stack and the lower-cost open-weight stack, and reshaping how AI agents and coding assistants are built and deployed.

via The New Stack|

#ai-pricing #coding-agents #deepseek

ai1 month ago•32 min saved

DeepSeek’s V4 AI preview targets US rivals with coding edge

Chinese AI firm DeepSeek released a preview of its V4 model, saying it can compete toe-to-toe with Google, OpenAI, and Anthropic, with a notable focus on coding. The open‑source approach and emphasis on compatibility with Huawei tech mark a milestone for China’s chip industry, but details on training costs and hardware remain undisclosed amid past accusations of banned Nvidia chips and alleged Claude misuse.

via The Verge|

#ai #coding #deepseek

technology1 month ago•312 min saved

DeepSeek’s V4 Preview Pushes China’s Open-Source AI Ambitions Forward

China’s DeepSeek unveiled a preview of its new V4 AI model, touting stronger reasoning and autonomous coding abilities, and powered by domestic Huawei/Cambricon chips to reduce Nvidia reliance. The open-source V4 aims to rival OpenAI, Anthropic and Google, continuing China’s push to scale AI applications across sectors. Analysts expect market reaction to be more muted than last year’s R1 debut, as the global AI landscape remains volatile amid US-China tech tensions over model distillation.

via CNN|

#china-ai #deepseek #huawei-ascend-950

technology1 month ago•2 min saved

DeepSeek Unveils V4-Pro and V4-Flash, Accelerating the US-China AI Showdown

DeepSeek released its latest chatbots, DeepSeek-V4-Pro and V4-Flash, claiming strong maths and coding performance and positioning them as open-source rivals to OpenAI and Google, with V4-Pro only slightly behind Google’s Gemini 3.1-Pro for knowledge and marginally trailing OpenAI’s GPT-5.4; the Flash version offers faster responses at lower cost. The rollout underscores the ongoing US–China AI race amid privacy and censorship concerns that led to bans on DeepSeek-R1 in several countries, while the Stanford AI Index 2026 notes China leads in publications, citations, patents, and industrial deployments despite the US producing more top-tier models.

via Al Jazeera|

#ai-models #artificial-intelligence #deepseek

technology3 months ago•6 min saved

Anthropic accuses Chinese labs of industrial-scale distillation attacks on Claude

Anthropic alleges that three Chinese AI labs—DeepSeek, Moonshot, and MiniMax—carried out industrial-scale distillation attacks to illicitly copy Claude, generating over 16 million exchanges with Claude through roughly 24,000 fraudulent accounts and violating terms of service and regional access restrictions; the company frames the campaigns as a national-security concern and is calling for rapid cross‑industry, government, and policy-maker cooperation to counter such abuses.

via Mashable|

#anthropic #claude #deepseek

technology3 months ago•1 min saved

Chinese AI Systems Tilt Responses Toward Beijing

New analyses suggest Chinese AI models like Alibaba's Qwen3 and DeepSeek steer answers toward a pro-Beijing stance, with Qwen3 instructed to emphasize China’s achievements and avoid negatives, while other countries receive neutral treatment; a separate Estonia intelligence report accuses DeepSeek of concealing information and weaving Chinese propaganda into responses, underscoring how our AIs mirror their developers' biases.

via Axios|

#ai-bias #chinese-ai #deepseek

business4 months ago•16 min saved

Nvidia Defies DeepSeek Panic as AI Pushes Valuation Toward $5 Trillion

One year after DeepSeek's hype wiped out about $589 billion from Nvidia's value, the stock has climbed roughly 58%, helping Nvidia approach a $5 trillion valuation as demand for high-end GPUs and AI data‑center infrastructure remains robust. Hyperscalers are continuing to invest about $600 billion in 2026 to expand data centers, broadening the AI hardware ecosystem to include memory makers and energy firms, while the China threat debate persists and U.S. tech leadership appears secure with rivals like Alphabet and Broadcom launching competing chips.

via TipRanks|

#ai #business #data-centers

technology4 months ago•26 min saved

DeepSeek’s Open-Weight Leap Prompts a China-Driven Reboot of the AI Arms Race

DeepSeek’s open-weight R1 stunned markets by matching elite LLMs at a fraction of the cost, triggering a $750 billion drop in the S&P 500 and a $590 billion hit to Nvidia on one day in 2025. Yet the story didn’t end there: U.S. hyperscalers plan over $600 billion in AI infrastructure for 2026, while Chinese firms push efficiency with open-weight models like Alibaba’s Qwen, seeding global adoption and capital inflows back to China. Critics say the real change is on the Chinese side, where cheaper, efficient architectures and state-driven AI policy are reshaping the race, even as the U.S. maintains a hardware edge and export controls. The long-run winner may hinge on whether scale or smarter design proves most valuable.

via MarketWatch|

#ai-infrastructure #chinese-ai #deepseek

business4 months ago•1 min saved

Chinese AI Companies' IPOs Boost Valuations and Market Presence

Shares of DeepSeek's rival doubled in their debut as Chinese AI companies rush to list on the stock market, reflecting strong investor interest in the sector.

via Financial Times|

#business #chinese-ai #deepseek

technology6 months ago•6 min saved

AI Chatbots Compared: The Unexpected Vacation Planning Winner

A travel writer tested five AI chatbots to plan a family trip to South Dakota, finding Deepseek the most effective for itinerary planning, while Google Gemini excelled at mapping, and others like ChatGPT and Microsoft Copilot were less practical.

via Tom's Guide|

#ai-chatbots #deepseek #google-gemini

technology7 months ago•4 min saved

Nvidia's Huang Criticizes AMD-OpenAI Deal Amid $1 Trillion AI Market Growth

Nvidia's Jensen Huang criticizes OpenAI's deal with AMD, highlighting industry shifts towards diversified AI hardware supply chains. Reflection AI raises $2 billion to develop AI training software, challenging the dominance of massive infrastructure investments. Meanwhile, Amazon's Prime Day savings are minimal, and AI chatbot usage among children is growing. Tesla unveils lower-cost models with downgrades, disappointing some analysts, as the EV market faces pricing and feature challenges.

via Sherwood News|

#ai-infrastructure #deepseek #nvidia

technology8 months ago•3 min saved

DeepSeek AI Model in China Cost $294,000 to Train, Developer Reveals

DeepSeek's reported $294,000 training cost is misleading; the actual cost to train their base model was around $5.87 million, with the lower figure referring only to a specific reinforcement learning phase, not the entire training process. The article clarifies misconceptions about the expenses involved in developing large AI models and compares DeepSeek's efforts to Western counterparts like Meta's Llama 4.

via theregister.com|

#ai-training-costs #deepseek #gpu-resources

technology8 months ago•3 min saved

China's DeepSeek AI Model Costs $294,000 to Train and Innovates Self-Reasoning

DeepSeek, a Chinese AI developer, revealed in a peer-reviewed article that it spent only $294,000 to train its R1 model using 512 Nvidia H800 chips, a significantly lower cost than US rivals, sparking renewed debate over China's role in the AI industry and raising questions about the technology and costs involved in AI development.

via CNN|

#ai-race #ai-training-costs #deepseek

technology8 months ago•2 min saved

DeepSeek AI Model's Cost and Innovation Unveiled in Landmark Paper

DeepSeek's R1 AI model, designed for reasoning tasks like math and coding, was trained cost-effectively at around $294,000 using reinforcement learning without copying from other LLMs, and has undergone peer review, setting a new standard for transparency and innovation in AI development.

via Nature|

#ai-model #deepseek #peer-review