home.social

#asic — Public Fediverse posts

Live and recent posts from across the Fediverse tagged #asic, aggregated by home.social.

fetched live
  1. Cash settlements on home insurance claims leaving vulnerable Australians short-changed, Asic says
    By Luca Ittimani

    Corporate regulator finds insurers settling most claims with cash rather than by managing repairs or rebuilding, warning of ‘unfair outcomes’ for homeowners

    theguardian.com/australia-news

    #InsuranceAustralia #Asic #Australianews #Business #Naturaldisasters #Consumeraffairs #TheGuardian #LucaIttimani

  2. Cash settlements on home insurance claims leaving vulnerable Australians short-changed, Asic says
    By Luca Ittimani

    Corporate regulator finds insurers settling most claims with cash rather than by managing repairs or rebuilding, warning of ‘unfair outcomes’ for homeowners

    theguardian.com/australia-news

    #InsuranceAustralia #Asic #Australianews #Business #Naturaldisasters #Consumeraffairs #TheGuardian #LucaIttimani

  3. Cash settlements on home insurance claims leaving vulnerable Australians short-changed, Asic says
    By Luca Ittimani

    Corporate regulator finds insurers settling most claims with cash rather than by managing repairs or rebuilding, warning of ‘unfair outcomes’ for homeowners

    theguardian.com/australia-news

    #InsuranceAustralia #Asic #Australianews #Business #Naturaldisasters #Consumeraffairs #TheGuardian #LucaIttimani

  4. #OpenAI’s #Jalapeño, a self-designed #ASIC for #LLMinference, outperforms #Nvidia’s #Blackwell and other competitors in #tokenthroughput per watt. Built with Broadcom, the chip uses HBM4 memory and a generalised architecture, excelling in both low-latency and high-throughput scenarios. While comparisons to Blackwell are incomplete, Jalapeño’s performance against Vera Rubin, which also uses HBM4, is noteworthy. newsletter.semianalysis.com/p/ #tech #news #ainews

  5. #OpenAI’s #Jalapeño, a self-designed #ASIC for #LLMinference, outperforms #Nvidia’s #Blackwell and other competitors in #tokenthroughput per watt. Built with Broadcom, the chip uses HBM4 memory and a generalised architecture, excelling in both low-latency and high-throughput scenarios. While comparisons to Blackwell are incomplete, Jalapeño’s performance against Vera Rubin, which also uses HBM4, is noteworthy. newsletter.semianalysis.com/p/ #tech #news #ainews

  6. #OpenAI’s #Jalapeño, a self-designed #ASIC for #LLMinference, outperforms #Nvidia’s #Blackwell and other competitors in #tokenthroughput per watt. Built with Broadcom, the chip uses HBM4 memory and a generalised architecture, excelling in both low-latency and high-throughput scenarios. While comparisons to Blackwell are incomplete, Jalapeño’s performance against Vera Rubin, which also uses HBM4, is noteworthy. newsletter.semianalysis.com/p/ #tech #news #ainews

  7. #OpenAI’s #Jalapeño, a self-designed #ASIC for #LLMinference, outperforms #Nvidia’s #Blackwell and other competitors in #tokenthroughput per watt. Built with Broadcom, the chip uses HBM4 memory and a generalised architecture, excelling in both low-latency and high-throughput scenarios. While comparisons to Blackwell are incomplete, Jalapeño’s performance against Vera Rubin, which also uses HBM4, is noteworthy. newsletter.semianalysis.com/p/ #tech #news #ainews

  8. #OpenAI’s #Jalapeño, a self-designed #ASIC for #LLMinference, outperforms #Nvidia’s #Blackwell and other competitors in #tokenthroughput per watt. Built with Broadcom, the chip uses HBM4 memory and a generalised architecture, excelling in both low-latency and high-throughput scenarios. While comparisons to Blackwell are incomplete, Jalapeño’s performance against Vera Rubin, which also uses HBM4, is noteworthy. newsletter.semianalysis.com/p/ #tech #news #ainews

  9. [Перевод] OpenAI Jalapeño: Персональный чип OpenAI, который обогнал NVIDIA Blackwell

    OpenAI создает свой чип Jalapeño: конец монополии Nvidia на инференс? SemiAnalysis выпустили разбор собственной аппаратной разработки OpenAI под кодовым названием Jalapeño. Пока все обсуждали закупки тысяч GPU Nvidia Blackwell, команда Сэма Альтмана совместно с Broadcom втайне проектировала специализированный ASIC на техпроцессе TSMC N3P, заточенный исключительно под инференс моделей масштаба GPT-4.5 и GPT-5.

    habr.com/ru/articles/1074526/

    #openai #jalapeno #nvidia #blackwell #semianalysis #asic #inference

  10. [Перевод] OpenAI Jalapeño: Персональный чип OpenAI, который обогнал NVIDIA Blackwell

    OpenAI создает свой чип Jalapeño: конец монополии Nvidia на инференс? SemiAnalysis выпустили разбор собственной аппаратной разработки OpenAI под кодовым названием Jalapeño. Пока все обсуждали закупки тысяч GPU Nvidia Blackwell, команда Сэма Альтмана совместно с Broadcom втайне проектировала специализированный ASIC на техпроцессе TSMC N3P, заточенный исключительно под инференс моделей масштаба GPT-4.5 и GPT-5.

    habr.com/ru/articles/1074526/

    #openai #jalapeno #nvidia #blackwell #semianalysis #asic #inference

  11. [Перевод] OpenAI Jalapeño: Персональный чип OpenAI, который обогнал NVIDIA Blackwell

    OpenAI создает свой чип Jalapeño: конец монополии Nvidia на инференс? SemiAnalysis выпустили разбор собственной аппаратной разработки OpenAI под кодовым названием Jalapeño. Пока все обсуждали закупки тысяч GPU Nvidia Blackwell, команда Сэма Альтмана совместно с Broadcom втайне проектировала специализированный ASIC на техпроцессе TSMC N3P, заточенный исключительно под инференс моделей масштаба GPT-4.5 и GPT-5.

    habr.com/ru/articles/1074526/

    #openai #jalapeno #nvidia #blackwell #semianalysis #asic #inference

  12. Broadcom’s AI Revenue Seen Reaching $116 Billion by FY2027 — Samsung Electronics Poised to Benefit in HBM and Foundry — BigGo Finance

    Broadcom’s artificial intelligence (AI) semiconductor business is expected to maintain its steep growth trajectory through fiscal 2027. As…
    #EuropeSays #Korea #KR #SamsungElectronics #ASIC #Broadcom #Citi #foundry #Google #HBM #JPMorgan #MarvellTechnology #Meta #Samsung
    europesays.com/korea/128869/

  13. AI deepfake scams ‘an emergency in the making’ as ASIC reports rise in false investment endorsements

    Scammers are using impersonations of celebrities, politicians and trusted public figures to promote fake investment schemes, with the…
    #NewsBeep #News #Artificialintelligence #AI #alankohler #ArtificialIntelligence #asic #AU #Australia #deepfake #fakenews #GinaRinehart #johnlaws #Online #scams #Technology
    newsbeep.com/au/857638/

  14. AI deepfake scams ‘an emergency in the making’ as ASIC reports rise in false investment endorsements

    Scammers are using impersonations of celebrities, politicians and trusted public figures to promote fake investment schemes, with the…
    #NewsBeep #News #Artificialintelligence #AI #alankohler #ArtificialIntelligence #asic #AU #Australia #deepfake #fakenews #GinaRinehart #johnlaws #Online #scams #Technology
    newsbeep.com/au/857638/

  15. Online trading platforms offer cash vouchers for ‘high-risk’ investments, watchdog ASIC says

    Retail investors are allegedly being offered cash vouchers and airline reward points as encouragement to use online platforms…
    #NewsBeep #News #Business #asic #AU #Australia #investments #investortrading #onlinetrading #tradingplatforms
    newsbeep.com/au/853340/

  16. Online trading platforms offer cash vouchers for ‘high-risk’ investments, watchdog ASIC says

    Retail investors are allegedly being offered cash vouchers and airline reward points as encouragement to use online platforms…
    #NewsBeep #News #Business #asic #AU #Australia #investments #investortrading #onlinetrading #tradingplatforms
    newsbeep.com/au/853340/

  17. I'm finally back home after an exciting HeiChips Summer School 2026. heichips.github.io/
    One week filled with loads of enthusiasm and energy for open-source chip design! 🎉

    The program includes lectures and workshops, as well as a final two-day hackathon, through which participants gained hands-on experience with the tools and PDK.

    #HeiChips26 #ASIC #OpenSource

  18. I'm finally back home after an exciting HeiChips Summer School 2026. heichips.github.io/
    One week filled with loads of enthusiasm and energy for open-source chip design! 🎉

    The program includes lectures and workshops, as well as a final two-day hackathon, through which participants gained hands-on experience with the tools and PDK.

  19. I'm finally back home after an exciting HeiChips Summer School 2026. heichips.github.io/
    One week filled with loads of enthusiasm and energy for open-source chip design! 🎉

    The program includes lectures and workshops, as well as a final two-day hackathon, through which participants gained hands-on experience with the tools and PDK.

    #HeiChips26 #ASIC #OpenSource

  20. I'm finally back home after an exciting HeiChips Summer School 2026. heichips.github.io/
    One week filled with loads of enthusiasm and energy for open-source chip design! 🎉

    The program includes lectures and workshops, as well as a final two-day hackathon, through which participants gained hands-on experience with the tools and PDK.

    #HeiChips26 #ASIC #OpenSource

  21. I'm finally back home after an exciting HeiChips Summer School 2026. heichips.github.io/
    One week filled with loads of enthusiasm and energy for open-source chip design! 🎉

    The program includes lectures and workshops, as well as a final two-day hackathon, through which participants gained hands-on experience with the tools and PDK.

    #HeiChips26 #ASIC #OpenSource

  22. RT @ns123abc: 🚨 EILMELDUNG: AMD hat gerade Taalas übernommen. Nvidia kaufte Groq für 20 Mrd. US-Dollar. SRAM-basierte LPU (Low-Power Units) sind modellagnostisch und bieten niedrige Latenzzeiten – sie können jedes Modell schnell ausführen. Jetzt kauft AMD Taalas: model-spezifische ASICs (Application-Specific Integrated Circuits). Das Modell wird in den Chip fest integriert – ein Modell, maximale Effizienz, null Flexibilität. Der Kampf um KI-Inferenz naht.

    mehr auf Arint.info

    #AMD #ASIC #Hardware #Inferenz #KI #Taalas #arint_info

    https://x.com/ns123abc/status/2085474824424493087#m

  23. RT @ns123abc: 🚨 EILMELDUNG: AMD hat gerade Taalas übernommen. Nvidia kaufte Groq für 20 Mrd. US-Dollar. SRAM-basierte LPU (Low-Power Units) sind modellagnostisch und bieten niedrige Latenzzeiten – sie können jedes Modell schnell ausführen. Jetzt kauft AMD Taalas: model-spezifische ASICs (Application-Specific Integrated Circuits). Das Modell wird in den Chip fest integriert – ein Modell, maximale Effizienz, null Flexibilität. Der Kampf um KI-Inferenz naht.

    mehr auf Arint.info

    #AMD #ASIC #Hardware #Inferenz #KI #Taalas #arint_info

    https://x.com/ns123abc/status/2085474824424493087#m

  24. RT @ns123abc: 🚨 EILMELDUNG: AMD hat gerade Taalas übernommen. Nvidia kaufte Groq für 20 Mrd. US-Dollar. SRAM-basierte LPU (Low-Power Units) sind modellagnostisch und bieten niedrige Latenzzeiten – sie können jedes Modell schnell ausführen. Jetzt kauft AMD Taalas: model-spezifische ASICs (Application-Specific Integrated Circuits). Das Modell wird in den Chip fest integriert – ein Modell, maximale Effizienz, null Flexibilität. Der Kampf um KI-Inferenz naht.

    mehr auf Arint.info

    #AMD #ASIC #Hardware #Inferenz #KI #Taalas #arint_info

    https://x.com/ns123abc/status/2085474824424493087#m

  25. Interesting move by AMD:
    servethehome.com/amd-to-acquir

    Basically yet another iteration of the "if software is too slow, make it hardware" continuum. This time a AMD acquisition turns LLMs into single-model-chips.

    Curious to see if and how this will play out to commercial usability and on what type of model.

    Also:

    • Might we even see coprocessor boards with exchangeable model chips?
    • What will the energy consumptions of those look like compared to the full generals purpose inference runner setups?

    #amd #hardware #servethehome #llm #customsilicon #ai #asic #innovation #mergersandacquisitions

  26. Interesting move by AMD:
    servethehome.com/amd-to-acquir

    Basically yet another iteration of the "if software is too slow, make it hardware" continuum. This time a AMD acquisition turns LLMs into single-model-chips.

    Curious to see if and how this will play out to commercial usability and on what type of model.

    Also:

    • Might we even see coprocessor boards with exchangeable model chips?
    • What will the energy consumptions of those look like compared to the full generals purpose inference runner setups?

    #amd #hardware #servethehome #llm #customsilicon #ai #asic #innovation #mergersandacquisitions

  27. Interesting move by AMD:
    servethehome.com/amd-to-acquir

    Basically yet another iteration of the "if software is too slow, make it hardware" continuum. This time a AMD acquisition turns LLMs into single-model-chips.

    Curious to see if and how this will play out to commercial usability and on what type of model.

    Also:

    • Might we even see coprocessor boards with exchangeable model chips?
    • What will the energy consumptions of those look like compared to the full generals purpose inference runner setups?

    #amd #hardware #servethehome #llm #customsilicon #ai #asic #innovation #mergersandacquisitions

  28. Interesting move by AMD:
    servethehome.com/amd-to-acquir

    Basically yet another iteration of the "if software is too slow, make it hardware" continuum. This time a AMD acquisition turns LLMs into single-model-chips.

    Curious to see if and how this will play out to commercial usability and on what type of model.

    Also:

    • Might we even see coprocessor boards with exchangeable model chips?
    • What will the energy consumptions of those look like compared to the full generals purpose inference runner setups?

    #amd #hardware #servethehome #llm #customsilicon #ai #asic #innovation #mergersandacquisitions

  29. Interesting move by AMD:
    servethehome.com/amd-to-acquir

    Basically yet another iteration of the "if software is too slow, make it hardware" continuum. This time a AMD acquisition turns LLMs into single-model-chips.

    Curious to see if and how this will play out to commercial usability and on what type of model.

    Also:

    • Might we even see coprocessor boards with exchangeable model chips?
    • What will the energy consumptions of those look like compared to the full generals purpose inference runner setups?

    #amd #hardware #servethehome #llm #customsilicon #ai #asic #innovation #mergersandacquisitions

  30. 📚🤔 Oh, sure! Let's just casually reverse-engineer an #ASIC in our spare time—because that's totally what normal people do, right? Let's just ignore the fact that it's like trying to unravel a ball of yarn that's been tangled by a hyperactive kitten. 🙄💻
    blog.janestreet.com/can-you-re #reverseengineering #hobbyprojects #techhumor #programming #challenges #hardwarehacking #HackerNews #ngated

  31. 📚🤔 Oh, sure! Let's just casually reverse-engineer an #ASIC in our spare time—because that's totally what normal people do, right? Let's just ignore the fact that it's like trying to unravel a ball of yarn that's been tangled by a hyperactive kitten. 🙄💻
    blog.janestreet.com/can-you-re #reverseengineering #hobbyprojects #techhumor #programming #challenges #hardwarehacking #HackerNews #ngated

  32. 📚🤔 Oh, sure! Let's just casually reverse-engineer an #ASIC in our spare time—because that's totally what normal people do, right? Let's just ignore the fact that it's like trying to unravel a ball of yarn that's been tangled by a hyperactive kitten. 🙄💻
    blog.janestreet.com/can-you-re #reverseengineering #hobbyprojects #techhumor #programming #challenges #hardwarehacking #HackerNews #ngated

  33. 📚🤔 Oh, sure! Let's just casually reverse-engineer an #ASIC in our spare time—because that's totally what normal people do, right? Let's just ignore the fact that it's like trying to unravel a ball of yarn that's been tangled by a hyperactive kitten. 🙄💻
    blog.janestreet.com/can-you-re #reverseengineering #hobbyprojects #techhumor #programming #challenges #hardwarehacking #HackerNews #ngated

  34. 📚🤔 Oh, sure! Let's just casually reverse-engineer an #ASIC in our spare time—because that's totally what normal people do, right? Let's just ignore the fact that it's like trying to unravel a ball of yarn that's been tangled by a hyperactive kitten. 🙄💻
    blog.janestreet.com/can-you-re #reverseengineering #hobbyprojects #techhumor #programming #challenges #hardwarehacking #HackerNews #ngated