/TechAnalysis argues Meta undercuts GPT-5.6 Sol by 86% with its $4.25 per million token Muse Spark 1.1 pricingFrançois Fleuret warns a monopolistic AI market poses severe societal risksFFJLT(AWCECombined views180K9 posts, first seen 51d ago642 likes49 commentsAnalysis argues Meta undercuts GPT-5.6 Sol by 86% with its $4.25 per million token Muse Spark 1.1 pricingFrançois Fleuret warns a monopolistic AI market poses severe societal risksFFJLT(AWCEReactions7 postsFFFrançois Fleuret@francoisfleuret7 weeks agoIt is hard to overstate how important it is for society at large that there are several heavyweights in competition in that space. One monopolistic actor as humankind enters the age of AI would be an absolute nightmare.… Show morelikes: 41replies: 2bookmarks: 6reposts: 6JLJason Lee@jasondeanlee7 weeks agoWhy do we need new closed model that's not better than open models?? https://twitter.com/ArtificialAnlys/status/2075677416295739660likes: 66replies: 10bookmarks: 7reposts: 0JLJason Lee@jasondeanlee7 weeks agoCan someone add glm 5.2 here? Suspect it's not even better than the open model https://twitter.com/alexandr_wang/status/2075218936266998230likes: 41replies: 3bookmarks: 7reposts: 0JLJason Lee@jasondeanlee7 weeks agoAt least grok 4.5 seem to be on the pareto frontier. Spark 1.1 is a new closed model off the frontier. Just shelve it instead of releasing? https://twitter.com/ArtificialAnlys/status/2075739735822307577likes: 21replies: 1bookmarks: 3reposts: 1T(Teortaxes▶️ (DeepSeek 推特🐋铁粉 2023 – ∞)@teortaxesTex7 weeks agoMuse Spark 1.1 is surprisingly close to Grok 4.5 on many high-signal evals This is the current top on CritPT https://twitter.com/ArtificialAnlys/status/2075677416295739660likes: 143replies: 6bookmarks: 13reposts: 4AWAlexandr Wang@alexandr_wang7 weeks agomuse spark can serve all your particle playground needs https://twitter.com/justpulket/status/2075826934177812537likes: 330replies: 27bookmarks: 41reposts: 21Show allCombined views180K9 posts, first seen 51d ago642 likes
FFFrançois Fleuret@francoisfleuret7 weeks agoIt is hard to overstate how important it is for society at large that there are several heavyweights in competition in that space. One monopolistic actor as humankind enters the age of AI would be an absolute nightmare.… Show morelikes: 41replies: 2bookmarks: 6reposts: 6
JLJason Lee@jasondeanlee7 weeks agoWhy do we need new closed model that's not better than open models?? https://twitter.com/ArtificialAnlys/status/2075677416295739660likes: 66replies: 10bookmarks: 7reposts: 0
JLJason Lee@jasondeanlee7 weeks agoCan someone add glm 5.2 here? Suspect it's not even better than the open model https://twitter.com/alexandr_wang/status/2075218936266998230likes: 41replies: 3bookmarks: 7reposts: 0
JLJason Lee@jasondeanlee7 weeks agoAt least grok 4.5 seem to be on the pareto frontier. Spark 1.1 is a new closed model off the frontier. Just shelve it instead of releasing? https://twitter.com/ArtificialAnlys/status/2075739735822307577likes: 21replies: 1bookmarks: 3reposts: 1
T(Teortaxes▶️ (DeepSeek 推特🐋铁粉 2023 – ∞)@teortaxesTex7 weeks agoMuse Spark 1.1 is surprisingly close to Grok 4.5 on many high-signal evals This is the current top on CritPT https://twitter.com/ArtificialAnlys/status/2075677416295739660likes: 143replies: 6bookmarks: 13reposts: 4
AWAlexandr Wang@alexandr_wang7 weeks agomuse spark can serve all your particle playground needs https://twitter.com/justpulket/status/2075826934177812537likes: 330replies: 27bookmarks: 41reposts: 21