Qwen 3.5: small models with impressive performancetwitter.com6 points·by moondistance·4 माह पहले·0 comments
Language Models Are Injective and Hence Invertiblearxiv.org1 points·by moondistance·9 माह पहले·2 comments
DeepSeek-R1 at 3,872 tokens / second on a single Nvidia HGX H200blogs.nvidia.com13 points·by moondistance·पिछला वर्ष·1 comments
ByteDance Doubao-1.5-pro matches GPT 4o benchmarks at 50x cheapertwitter.com4 points·by moondistance·पिछला वर्ष·0 comments
US-China Commission top recommendation: Manhattan project for race to AGI [pdf]uscc.gov4 points·by moondistance·2 वर्ष पहले·0 comments
AI scans RNA 'dark matter' and uncovers 70k new virusesnature.com1 points·by moondistance·2 वर्ष पहले·0 comments
moondistance·11 माह पहले·discussHow does this compare with Exowatt? Assume fresnel lenses are more efficient.
moondistance·पिछला वर्ष·discussChris Crawford is also famous for the Dragon Speech :) https://youtu.be/CBrj4S24074?si=Ph12RpW8BKsh8-qS
moondistance·पिछला वर्ष·discussYes, big time, but there continues to be lots of progress.Most importantly, models are maturing, and this means less custom optimization is required.
moondistance·पिछला वर्ष·discuss(this is also why AMD popped 10% at open yesterday - this is a new development and talks from their 2025 "Advancing AI" event were published late last week + over the weekend)
moondistance·पिछला वर्ष·discussYes, for many applications.Meta, OpenAI, Crusoe, and xAI recently announced large purchases of MI300 chips for inference.MI400, which will be available next year, also looks to be at least on par with Nvidia's roadmap.
moondistance·पिछला वर्ष·discussWhere do you find/buy land? How do you vet purchases? Can you point to a few websites, etc.? Thanks!
moondistance·2 वर्ष पहले·discussHappy New Year!I know so few of you but feel like you’re my best friends. Love to all.
moondistance·2 वर्ष पहले·discussMicrosoft is negotiating equity in OpenAI as part of the switch to a for-profit. Non-zero chance this is a negotiation flex.