Inception launches Mercury 3, a diffusion-based LLM it claims is
Inception launches Mercury 3, a diffusion-based LLM it claims is "comparable to Luna and Haiku models" but 4x to 8x faster, generating ~1,000 tokens per second
TL;DRNew diffusion model generates text 4-8x faster than comparable LLMs.
Why it matters: Speed gains could reduce inference costs and enable real-time AI applications at scale.
At Apple, Pigs Are About to Fly
TL;DRApple is announcing new MacBook Pro models next week.
Cloudflare debuts Clairic, a System One-class extraction model it says
TL;DRCloudflare built a specialized AI model for data extraction that's faster and cheaper than general LLMs.
Latest
Korean data-center company Elice falls 30%+ in Kosdaq trading debut;
South Korean data-center firm Elice tanked 30% on IPO debut amid sector concerns.
OpenAI is racing to secure investor commitments before Anthropic's
OpenAI chasing $30B funding round before Anthropic goes public, seeking $1.4T+ valuation.
Mamdani says feds have abdicated on AI, urges New Yorkers to push back
NYC mayor says feds aren't regulating AI; locals must demand protections.
Citi Global Chief Economist on Global Growth
Citi's chief economist flags concerns about energy costs, inflation, and AI disruption affecting global growth.
Citi Australia CEO on ANZ Market, Biz Growth
Citi's Australia CEO sees strong M&A activity and growth opportunities in the region.
Interpretability Is the Key to Physical AI, Says World Mechanics CEO
Making AI understand physics requires knowing how AI actually thinks internally.
How to stop audio from auto-playing on Android Auto
Disable autoplay in Android Auto settings to stop music playing automatically.
Quminex raises $1.5M CAD ($1.06M USD) Pre-Seed Round Led by Quantacet
Quebec mining tech startup Quminex raised $1.06M to develop mineral discovery platform.
Palo Alto Networks CEO Advises Daughter on New AI-Focused Venture
Palo Alto Networks CEO's daughter launching AI-focused VC fund with insider guidance.