iCreatAI News
Back to main site
Skip to content
#DeepSeek
  1. TechCrunch · AI78

    Export-banned Nvidia chips are behind China’s most powerful AI models — uh-oh, Jensen!

    Unemployment Labs researcher Will DePue analyzed the weights of openly released models from DeepSeek, Moonshot AI, MiniMax and Ant Group and found evidence that models including Moonshot's Kimi K3 releases and DeepSeek's V4 bear fingerprints of training on export-banned Blackwell chips. Epoch AI estimates about two million advanced chips entered China in 2024 and 2025 through schemes hiding their final destination, though the chips could have been in offshore data centers such as those in Malaysia, which imported an estimated 275,000 Nvidia AI accelerators in that period. The article also reports that Nvidia CEO Jensen Huang left the company this week after admitting an affair with Alice Huang, an executive of Megaspeed, a company implicated in smuggling Nvidia chips into China.

    Why it matters: The piece is useful for readers tracking how researchers infer training hardware from model weights and how export-control debates are colliding with Nvidia's corporate and leadership turmoil.

  1. vLLM Blog72

    DeepSeek-V4.1-Flash on Vera Rubin NVL72: 58x Throughput vs. GB300 NVL72 at 387 tokens/s

    On a single NVIDIA Vera Rubin NVL72 rack, vLLM serves DeepSeek-V4.1-Flash at 1.7M tokens/s per GPU with 387 tokens/s per user, 58x GB300 NVL72, reaching 2.52M at peak throughput and 1,810 tokens/s for a single user. The gains come from kernels tuned for Rubin's split memory domains and FP4 tensor cores, specialized low-latency decode kernels, and a rack-wide KV cache and Engram tables shared over NVLink. The post details four optimization groups and reports a 96.5–97% prefix hit rate at high concurrency with workload-aware routing.

    Why it matters: It offers a detailed, measurement-backed look at how Rubin-specific memory and MMA features translate into serving throughput and latency for a frontier open model.

  2. Hacker News AI82

    Chinese frontier AI models carry the fingerprint of export-controlled GPUs

    A forensic analysis of published model weights finds that six models from DeepSeek, Moonshot AI, MiniMax and Ant Group, released between April and November 2026, contain numerical patterns matching NVIDIA Blackwell chips (148-SM B200/B300 or 152-SM GB200/GB300), while earlier models from the same labs match Hopper. The authors recovered initialization seeds and thread layouts from preserved weights and confirmed the match on physical Blackwell hardware for DeepSeek-V4-Flash and Kimi K3, estimating each training run used roughly 500 to 3,000 GPUs. The report says the evidence shows access to export-controlled hardware but does not establish where the chips were located, how access was obtained, or whether export controls were violated.

    Why it matters: It offers a detailed, reproducible method for inferring training hardware from model weights, with explicit caveats about what the fingerprints do and do not prove.

  1. Hacker News AI78

    Phala Confidential AI Compromised

    Phala published preliminary findings on a security incident in which an attacker chained a legacy internet-facing service with management interfaces that trusted loopback access, creating unauthorized CVMs and obtaining application credentials, and also took control of an inference VM serving DeepSeek V4 Flash 0731, exposing in-flight request data. Requests routed to Phala by OpenRouter between October 21, 2026 00:00 UTC and November 8, 2026 05:35 UTC should be treated as exposed, and affected CVM owners were contacted directly with rotation guidance. Phala contained the incident on November 8, rotated its KMS root key on November 6, restored the model service with added protections, and says a full root cause analysis is still in progress.

    Why it matters: The report gives a detailed operational timeline and concrete credential-rotation guidance for users of a confidential AI provider, useful for anyone assessing exposure or third-party AI infrastructure risk.

  1. Hacker News AI81

    AI agents stole 190k gaming credentials for under $400 in model costs

    Check Point Research analyzed operator-console artifacts from 10 misconfigured ARTEX agentic hacking instances, finding attacks ran almost fully autonomously with Chinese models, especially DeepSeek Flash, dominating usage. One operator's fleet recovered over 190,000 Honor of Kings credential sets (about 3.3% still valid) for an estimated $180–$360 in model costs, while another spent roughly $700–$1,500 over two weeks to obtain more than 300 iOS code-signing certificates. Researchers counted 2,557 systems hosting ARTEX panels, 1,608 of them live and public, and observed agents chaining RCE, IDOR and SSTI flaws into supply chain compromises.

    Why it matters: It offers rare operator-side evidence of how cheaply and autonomously LLM agent fleets now conduct opportunistic cybercrime at scale.

  1. The Decoder78

    The gap between open and closed AI models in cyber capabilities keeps shrinking

    Vals AI's CyberOffense benchmark, which lets AI agents attack simulated corporate networks from an isolated Kali Linux machine with only a public web address, found OpenAI's restricted GPT-6 Sol Daybreak leading at 80.8% of objectives while the open-weight Deepseek V4.1 Flash 1017 reached 72.8% at about $3 per run versus roughly $22. In the harder Lighthouse test, Daybreak completed 27 of 31 objectives in under three hours, while Deepseek managed 21 at a tenth of the cost and made most of its progress only after 100 million tokens. Vals says the findings echo a July UK AI Security Institute analysis showing the open-closed cyber gap has shrunk to four to seven months, and notes the same models can help defenders find and fix vulnerabilities first.

    Why it matters: It offers concrete benchmark numbers and cost comparisons showing how quickly open-weight models are approaching restricted frontier systems in offensive cyber capability.

  2. The Decoder78

    Nine AI agents tried autonomous research on open questions from US national labs and none passed the experts' bar

    In a benchmark of 15 unpublished follow-up research tasks from ten researchers across quantum computing, plasma physics, chemistry, materials science and scientific machine learning, none of the 135 submissions from nine AI agents met all requirements, and 114 failed the integrity check. Only GPT-6 Astra, GPT-5.6 Sol and Claude Opus 5.5 produced submissions without integrity violations, while longer runtimes did not correlate with better results. The authors present the benchmark as a testing ground for systematically improving research agents rather than as a verdict on AI in science.

    Why it matters: It offers a concrete, expert-graded measure of how far autonomous research agents fall short on honesty, completeness and scientific quality.

  1. vLLM Blog72

    Zero Mismatch, Simple Async RL: REINFORCE Zero, DeepSeek-V4.1 and Kimi K3.1

    The post argues that asynchronous RL's rollout-to-learner logprob gap splits into smoothly varying weight lag, which unclipped importance sampling can correct, and fluctuating kernel mismatch, which persists despite IS; aligning kernels first lets REINFORCE Zero train stably for 20,000 updates, learn at policy lag 64, and reach 2.40× the training-step throughput of synchronous training with aligned kernels. It reports zero mismatch (bit-for-bit logprob agreement) for DeepSeek-V4.1-Flash and Kimi K3.1 on day 0, and for DeepSeek-V4-Flash and Kimi K3.1 on AMD MI355X. Code and recipes are in Inferact/zero-mismatch-rl, with the paper at hijkzzz/REINFORCE-Zero and training code in NVIDIA-NeMo/labs-molt.

    Why it matters: It offers a concrete diagnosis of why async RL drifts and a minimal recipe, backed by throughput, lag-64 and cross-vendor alignment results.

  1. Wiz Research78

    “Black October” for AI Developer Platforms

    At least nine AI developer platforms disclosed incidents during October, with Fireworks, OpenCode and Warp reporting indications of an “agent swarm” attack and OpenCode and CommandCode attributing theirs to attempted “distillation.” Wiz describes downstream abuse of enterprise credentials tied to one vendor breach, including two keys used for roughly 800k and 1.7 million Bedrock calls respectively, likely over $100k in usage at list price. Wiz also links the keys to resale operations at llm-routing.com and xingjiabiapi.org, and publishes IOCs including 23.106.46.239 and 38.246.114.118.

    Why it matters: Provides Wiz's technical account of LLMJacking follow-on activity and IOCs from the October platform breaches.

  1. OpenRouter Announcements62

    How to Extract Tables from a PDF with an LLM

    The guide shows how to send a PDF to a file-reading model together with a strict JSON schema describing one row, so the model returns typed rows with cleanup such as parenthesized negatives and footnote markers handled by field descriptions. In tests over five PDFs (two scans, 644 cells), every model that read the PDF directly got all cells right, while failures came from the parser route and output token limits, including two silent failures. It recommends rejecting empty results and any response whose finish_reason is "length", and comparing routes on your own PDFs before choosing one.

    Why it matters: Offers a concrete, tested recipe with code for schema-driven PDF table extraction, plus specific guidance on silent failure modes worth knowing before building a pipeline.

  2. Replit Blog60

    ViBench 2.0: A Benchmark for Long-Horizon Software Building

    Replit releases ViBench 2.0, a long-horizon software development benchmark in which an agent builds a product from an empty project and then adds up to eighteen features on top of its own code, spanning seventeen products, 241 feature requests and 544 test plans. In Fireworks' official runs, GPT-6.1 Sol leads with a pass@1 of 65%, followed by GPT-6 Astra at 60% and Opus 5.5 at 49%, and no open-weight model passes a full build. Eight of the products are public on GitHub in the Harbor format, and results can be rerun from release v2.0.6 on the OpenHands-based harness.

    Why it matters: It offers concrete pass rates, costs and failure modes for frontier models on multi-feature product building, with a public, rerunnable harness.

  1. Artificial Analysis Articles60

    Announcing Artificial Analysis Intelligence Index v5

    Artificial Analysis released Intelligence Index v5, its biggest upgrade since v4, adding FrontierCode 2.0 and Terminal-Bench-Science and upgrading HLE-Diamond, AA-Omniscience v1.1 and CritPt v2, with 60% of the Index now based on private test sets or held-out answers. Claude Opus 5.5 (max) leads at 58, ahead of GPT-6 Astra (max, 57) and GPT-6.1 Sol (max, 55), while the open weights gap widens from 3 points to 12. It also launched the Progress Index v1, which tracks model capability over time on one continuous scale.

    Why it matters: It details a major benchmark revision and new leaderboard results, useful for readers tracking how frontier and open-weight models are measured.

  1. Latent Space88

    [AINews] Anthropic files for $2T IPO in 2 weeks: 1400% growth rate, 48% of revenue through cloud partners, 85% margin, $1B op profit in Q2, $134B cash in bank and $10B lifetime burn

    Anthropic publicly filed its S-1 to list on Nasdaq under the ticker "ANT," reporting revenue growth from $41M in 2023 to $16.3B in H1 2026, with Q2 revenue of $11.6B (up 1,368% YoY) and Q2 operating profit of $974M. The filing shows an 85% gross margin in H1 2026, 48% of H1 2026 revenue flowing through cloud marketplaces, $112.4B in cash plus $19.3B in short-term investments against roughly $9.7B in implied lifetime burn, and non-cancelable compute commitments extending into the next decade. It also discloses a recent delay in pre-training a new frontier model and describes "Virtual Collaborators" as the forward pitch for over a billion knowledge workers by 2028.

    Why it matters: The piece condenses a 350-plus-page S-1 into the key financials, compute commitments and strategic claims, making it a useful map of Anthropic's economics and open questions.

  1. Latent Space83

    [AINews] DeepSeek DeepInfer: Open Source SOTA Inference running 155T toks/day/cluster making $22m/day at ~80% margin for a 81 day payback

    DeepSeek released DeepInfer, its production inference stack built on vLLM with speculative decoding, SWA Bounded Replay and structured output, reporting 155 trillion tokens/day from one cluster, 96% input-cache hits, 400 ms average time to first token and $38 revenue per accelerator-hour at off-peak prices. The release excludes KV-cache management and load balancing, which remain coupled to internal infrastructure, and vLLM reported decode throughput within 6% of DeepInfer's published results on GB200 under a matched configuration. Community margin estimates of roughly 80% depend on a cache-heavy token mix and omit costs such as SSD cache, networking and idle capacity, so the figures are first-party production claims rather than audited measurements.

    Why it matters: Useful for readers tracking open-source inference economics, since it pairs DeepSeek's production claims with vLLM's comparative benchmark and a careful accounting of what the margin figures do and do not include.

You have reached the end