CompanyAnthropic8 recent entries28 Aug 2026Double standard @sama . Exactly two years ago, he belittled @dylan522p as the 'SemiAnalysis' guy after not liking a Google analysis. Altman'…Double standard @sama . Exactly two years ago, he belittled @dylan522p as the 'SemiAnalysis' guy after not liking a Google analysis. Altman's now kissing Dylan's ass, inviting SemiAnalysis to OpenAI l→4 Sept 2026Been trying Anthropic Fable 5.1 vs OpenAI Astra recently. Pretty clear who the real winner is.The video, posted on September 4, 2026 by user Dylan Patel, compares Anthropic’s Fable 5.1 with OpenAI's Astra and concludes that one clearly outperforms the other. Filmed in Everett, Massachusetts, i
CompanyOpenAI8 recent entries7 Sept 2026TPU Inference Externalization Full Steam Ahead - InferenceXGoogle has begun allowing third‑party customers to acquire or rent its TPU chips—starting with the TPUv7 “Ironwood” generation. In comparative tests against NVIDIA Volta (B200/B300), Ironwood shows up→11 Sept 2026Nvidia’s Backstop Universe – Heads I Win, Tails Who Loses?The article examines Nvidia’s role in the 11 trillion AI infrastructure boom, exploring how the company finances this growth through AI Compute, Capital and Markets models that reveal funding sources,→26 Sept 2026Rubin HBM Despec = Nvidia cutting cokeRubin HBM Despec = Nvidia cutting coke Fentanyl Grade Compute: Why a GB300 Rack Will Out-Price Blow by Weight in 2030 A GB300 NVL72 weighs roughly 1,580 kg fully populated and recent purchase orders p→28 Sept 2026Samsung commits 1B to AI infrastructure company Helix, adding to the 10B already secured when a KKR-led consortium including Nvidia established Helix in June (Wall Street Journal)Wall Street Journal: Samsung commits 1B to AI infrastructure company Helix, adding to the 10B already secured when a KKR-led consortium including Nvidia established Helix in June — Helix expects to le→4 Oct 2026Q&A with OpenAI VP of Hardware Richard Ho on its Jalapeño inference chip co-designed with Broadcom, using internal OpenAI models to design the chip, and more (Dr. Ian Cutress/More Than Moore)Dr. Ian Cutress / More Than Moore: Q&A with OpenAI VP of Hardware Richard Ho on its Jalapeño inference chip co-designed with Broadcom, using internal OpenAI models to design the chip, and more — VP Ha→5 Oct 2026Ghost, which makes a 3,499 computer designed for AI agents and includes an RTX Pro 4000 SFF Blackwell GPU, emerges from stealth with an 11M seed led by a16z (Dominic-Madori Davis/TechCrunch)Dominic-Madori Davis / TechCrunch: Ghost, which makes a 3,499 computer designed for AI agents and includes an RTX Pro 4000 SFF Blackwell GPU, emerges from stealth with an 11M seed led by a16z — For mu→5 Oct 2026Anthropic Subscriptions Offer 5x+ More Value Than OpenAIAnthropic’s paid subscription plans deliver more than five‑times the perceived value of comparable OpenAI offerings when evaluated by cost per inference unit and overall revenue contribution. The heav→5 Oct 2026Analysis: Anthropic's subscriptions offer ~5x more API-equivalent value per month than OpenAI's for agentic workloads with Claude Opus 5.5 vs. GPT-6.1 Sol (SemiAnalysis)SemiAnalysis: Analysis: Anthropic's subscriptions offer ~5x more API-equivalent value per month than OpenAI's for agentic workloads with Claude Opus 5.5 vs. GPT-6.1 Sol — Limit testing every AI subscr
CompanyGoogle6 recent entries11 Apr 2026All ToolsAuto-generated index of all tools mentioned across the wiki.→11 Apr 2026All PeopleAuto-generated index of all people mentioned across the wiki.→11 Apr 2026All CompaniesAuto-generated index of all companies mentioned across the wiki.→19 Apr 2026A profile of Maria Davidson, who heads California Renewal, a pro-business political group backed by Silicon Valley power players, seeking to raise $100M in 2026 (Emily Shugerman/The San Francisco ...)Emily Shugerman / The San Francisco Standard: A profile of Maria Davidson, who heads California Renewal, a pro-business political group backed by Silicon Valley power players, seeking to raise $100M i→19 Jul 2026Wiki Lint Report — 2026-07-19Automated lint: 20 errors, 8743 warnings, 3 info→9 Aug 2026Google's AI shakeup suggests it may be prioritizing AI diffusion over frontier-model leadership, betting on AI compute as a bigger economic opportunity (Tim O'Reilly/Asimov's Addendum)Tim O'Reilly / Asimov's Addendum: Google's AI shakeup suggests it may be prioritizing AI diffusion over frontier-model leadership, betting on AI compute as a bigger economic opportunity — SemiAnalysis
CompanyMeta8 recent entries11 Apr 2026All CompaniesAuto-generated index of all companies mentioned across the wiki.→15 Apr 2026Social media intern made me a terminally online LEGO 😭Dylan Patel, a semiconductor and AI industry analyst known for his work at SemiAnalysis, shared a post on X (formerly Twitter) humorously reacting to a LEGO-style avatar or depiction of himself create→23 Apr 2026Every conversation I have with @dylan522p, I'm really just trying to understand the supply and demand of tokens. This is a unique episode in…Every conversation I have with @dylan522p, I'm really just trying to understand the supply and demand of tokens. This is a unique episode in that it's entirely dedicated to talking about both sides of→2 Jul 2026Meta Compute: Everyone Wants To Be A CloudMeta Compute explores Meta's strategy to position itself as a cloud infrastructure provider, leveraging its massive internal compute investments and AI capabilities to offer services to external custo→9 Jul 2026The Future of Meta Superintelligence: A 1 Year Progress UpdateMeta has made significant progress on its superintelligence research agenda, which aims to develop advanced AI systems with broad capabilities. The update covers Meta's infrastructure investments, res→19 Jul 2026Wiki Lint Report — 2026-07-19Automated lint: 20 errors, 8743 warnings, 3 info→28 Jul 2026🦔The cost of insuring Big Tech debt against default just hit record highs. Credit default swaps are insurance policies investors buy when t…🦔The cost of insuring Big Tech debt against default just hit record highs. Credit default swaps are insurance policies investors buy when they think a company might not pay back what it borrowed. When→5 Oct 2026Analysis: Anthropic's subscriptions offer ~5x more API-equivalent value per month than OpenAI's for agentic workloads with Claude Opus 5.5 vs. GPT-6.1 Sol (SemiAnalysis)SemiAnalysis: Analysis: Anthropic's subscriptions offer ~5x more API-equivalent value per month than OpenAI's for agentic workloads with Claude Opus 5.5 vs. GPT-6.1 Sol — Limit testing every AI subscr
CompanyMistral5 recent entries11 Apr 2026All ToolsAuto-generated index of all tools mentioned across the wiki.→11 Apr 2026All PeopleAuto-generated index of all people mentioned across the wiki.→11 Apr 2026All CompaniesAuto-generated index of all companies mentioned across the wiki.→19 Jul 2026Wiki Lint Report — 2026-07-19Automated lint: 20 errors, 8743 warnings, 3 info→8 Sept 2026Led by @Samsung, co-led by @eqt's Scaleup Europe Fund and @PSG_equity, with continued backing from @ASMLcompany, @nvidia, @BNPParibas CIB an…Led by @Samsung, co-led by @eqt's Scaleup Europe Fund and @PSG_equity, with continued backing from @ASMLcompany, @nvidia, @BNPParibas CIB and others. This fuels our infrastructure and product capabili
CompanyxAI4 recent entries11 Apr 2026All ToolsAuto-generated index of all tools mentioned across the wiki.→11 Apr 2026All PeopleAuto-generated index of all people mentioned across the wiki.→11 Apr 2026All CompaniesAuto-generated index of all companies mentioned across the wiki.→19 Jul 2026Wiki Lint Report — 2026-07-19Automated lint: 20 errors, 8743 warnings, 3 info
CompanyDeepSeek8 recent entries11 Apr 2026All CompaniesAuto-generated index of all companies mentioned across the wiki.→27 Apr 2026The DeepSeek V4 garbled output bug in open source inference engine is fixed in SGLang. To everyone affected over the weekend, sorry for the …The DeepSeek V4 garbled output bug in open source inference engine is fixed in SGLang. To everyone affected over the weekend, sorry for the trouble. Huge thanks to @Ant_Group for landing the fix PR. I→9 Jun 2026DeepSeekV4 1.6T Day 0 to Day 43 Performance Over Time - Huawei, GB300 NVL72, MI355X, B200This article from SemiAnalysis tracks the performance evolution of DeepSeek V4, a 1.6 trillion parameter model, over its first 43 days of training on Huawei and NVIDIA hardware including GB300 NVL72, →19 Jul 2026Wiki Lint Report — 2026-07-19Automated lint: 20 errors, 8743 warnings, 3 info→23 Jul 2026Vera Rubin NVL72 vs GB200 NVL72? Inference TCO & Architecture AnalysisVera Rubin NVL72 is Nvidia’s second‑generation, rack‑scale Oberon architecture that achieves inference gains through extreme co‑design. Early engineering‑sample data from CoreWeave show DeepSeek R1 de→25 Aug 2026OpenAI Jalapeño Inference Chip: First Measured ResultsOpenAI reports Jalapeño delivers 1.5–1.9× more AI work per watt and 1.7–3.6× lower latency vs comparison systems on InferenceX.→14 Sept 2026Vera Rubin NVL72 inference tests show up to 7x better token throughput per MW vs. Blackwell on a 1.6T DeepSeek model, above Huang's 3x claim for 1T-3T LLMs (Bryan Shan/SemiAnalysis)Bryan Shan / SemiAnalysis: Vera Rubin NVL72 inference tests show up to 7x better token throughput per MW vs. Blackwell on a 1.6T DeepSeek model, above Huang's 3x claim for 1T-3T LLMs — Jensen Sandbagg→18 Sept 2026Engrams Embedding Entendre: Codesign for Efficient DRAM/SSD OffloadingEngram Embedding Entendre is a model‑architecture optimization that extends standard token embeddings with learned multi‑token lookups, enabling recurring local patterns to be retrieved directly from
CompanyNVIDIA8 recent entries8 Sept 2026Led by @Samsung, co-led by @eqt's Scaleup Europe Fund and @PSG_equity, with continued backing from @ASMLcompany, @nvidia, @BNPParibas CIB an…Led by @Samsung, co-led by @eqt's Scaleup Europe Fund and @PSG_equity, with continued backing from @ASMLcompany, @nvidia, @BNPParibas CIB and others. This fuels our infrastructure and product capabili→11 Sept 2026Nvidia’s Backstop Universe – Heads I Win, Tails Who Loses?The article examines Nvidia’s role in the 11 trillion AI infrastructure boom, exploring how the company finances this growth through AI Compute, Capital and Markets models that reveal funding sources,→14 Sept 2026Vera Rubin NVL72 Agentic Inference: 67x better Performance per Dollar**Summary for AI knowledge base** Vera Rubin NVL72 is a co‑designed platform combining the Rubin GPU, Vera CPU, NVLink 6 Switch, ConnectX‑9, BlueField‑4, and Spectrum‑6, targeting agentic inference wo→18 Sept 2026Engrams Embedding Entendre: Codesign for Efficient DRAM/SSD OffloadingEngram Embedding Entendre is a model‑architecture optimization that extends standard token embeddings with learned multi‑token lookups, enabling recurring local patterns to be retrieved directly from →26 Sept 2026Rubin HBM Despec = Nvidia cutting cokeRubin HBM Despec = Nvidia cutting coke Fentanyl Grade Compute: Why a GB300 Rack Will Out-Price Blow by Weight in 2030 A GB300 NVL72 weighs roughly 1,580 kg fully populated and recent purchase orders p→28 Sept 2026Samsung commits 1B to AI infrastructure company Helix, adding to the 10B already secured when a KKR-led consortium including Nvidia established Helix in June (Wall Street Journal)Wall Street Journal: Samsung commits 1B to AI infrastructure company Helix, adding to the 10B already secured when a KKR-led consortium including Nvidia established Helix in June — Helix expects to le→4 Oct 2026Q&A with OpenAI VP of Hardware Richard Ho on its Jalapeño inference chip co-designed with Broadcom, using internal OpenAI models to design the chip, and more (Dr. Ian Cutress/More Than Moore)Dr. Ian Cutress / More Than Moore: Q&A with OpenAI VP of Hardware Richard Ho on its Jalapeño inference chip co-designed with Broadcom, using internal OpenAI models to design the chip, and more — VP Ha→5 Oct 2026Ghost, which makes a 3,499 computer designed for AI agents and includes an RTX Pro 4000 SFF Blackwell GPU, emerges from stealth with an 11M seed led by a16z (Dominic-Madori Davis/TechCrunch)Dominic-Madori Davis / TechCrunch: Ghost, which makes a 3,499 computer designed for AI agents and includes an RTX Pro 4000 SFF Blackwell GPU, emerges from stealth with an 11M seed led by a16z — For mu