Mindreading is coming, what’s the plan?20m ago India sees $11–12 bn chip investment interest under Semicon 2.0: Ashwini Vaishnaw20m ago Inside the suddenly explosive world of AI safety29m ago EU-backed Scaleup Europe Fund in talks to invest in ElevenLabs34m ago ‘Can we work out a deal where you pay ME and I get no AI at all?’ Meta wants you to pay for more AI — one brutal response suggests people might rather pay to make it go away36m ago Amodei Cites Recursive Self-Improvement In September Essay43m ago Your Employees Are Trusting AI With More Than You Know: Here’s How Governance Must Evolve44m ago Why Infrastructure AI — Not Consumer AI — Will Define the Next Decade55m ago EU’s Slow Regulation Is Finding Fans in AI Age57m ago Feeling overwhelmed by the AI doom loop? Here’s the essential reading list to make sense of it all58m ago Labor crisis in U.S. chip manufacturing leaves Samsung, TSMC desperate for talent59m ago Adecco Group rolls out Agentforce Coworker to 27,000 staff in 40-plus countries59m ago Japan and the US are discussing a GlobalFoundries chip plant under their $550bn pact, Nikkei reports1h ago OpenAI reports 6 new instances of 'concerning model behavior' since March1h ago

Mindreading is coming, what’s the plan?20m ago India sees $11–12 bn chip investment interest under Semicon 2.0: Ashwini Vaishnaw20m ago Inside the suddenly explosive world of AI safety29m ago EU-backed Scaleup Europe Fund in talks to invest in ElevenLabs34m ago ‘Can we work out a deal where you pay ME and I get no AI at all?’ Meta wants you to pay for more AI — one brutal response suggests people might rather pay to make it go away36m ago Amodei Cites Recursive Self-Improvement In September Essay43m ago Your Employees Are Trusting AI With More Than You Know: Here’s How Governance Must Evolve44m ago Why Infrastructure AI — Not Consumer AI — Will Define the Next Decade55m ago EU’s Slow Regulation Is Finding Fans in AI Age57m ago Feeling overwhelmed by the AI doom loop? Here’s the essential reading list to make sense of it all58m ago Labor crisis in U.S. chip manufacturing leaves Samsung, TSMC desperate for talent59m ago Adecco Group rolls out Agentforce Coworker to 27,000 staff in 40-plus countries59m ago Japan and the US are discussing a GlobalFoundries chip plant under their $550bn pact, Nikkei reports1h ago OpenAI reports 6 new instances of 'concerning model behavior' since March1h ago

Chips & Compute Machine Learning5 min reading time

Z.ai Details GLM-5.3-Flash Inference Build on 100,000 Chinese Chips

Unite.AIPublished 2 hours ago

Z.ai built a production inference system for its GLM-5.3-Flash model using over 100,000 Chinese AI chips, overcoming hardware and ecosystem challenges. The model supports a 1M-token context window and multimodal inputs, and Z.ai developed a dense feedback method to diagnose system issues efficiently.

#ai-infrastructure #glm-5-3-flash #z-ai

More in Chips & Compute

Chips & Compute

Huawei’s Plan to Become China’s Nvidia

The Wall Street Journal2h ago

Chips & Compute

The Startup That Built OpenAI’s Biggest Data Center Is Now Making Tiny Ones

The Wall Street Journal3h ago

Chips & Compute2 min read

Apple planning to sell AI servers powered by M8 Ultra chips, says report

Covered by 4 sources22h ago