Alibaba plans 5-10 trillion parameter Qwen models as Qwen 4 enters training
emmanuelvivier · x · 2026-09-23
Alibaba says it wants to train Qwen models at 5 to 10 trillion parameters, with Qwen 4 already in training. The company also unveiled the Zhenwu V900 chip, claimed to be China's most powerful AI chip, designed for clusters of 500,000 cards.
More from Infra
- Hugging Face TRL Gets Major Speedup: AsyncGRPO Training Steps Up to 3.5x Faster — ben_burtenshaw · 2026-09-23
- Qualcomm ships two mobile chips with Hexagon NPU built for on-device MoE AI agents — emmanuelvivier · 2026-09-23
- Go.AI raises $85M for on-prem AI infrastructure serving banks and healthcare — emmanuelvivier · 2026-09-23
- Crusoe signs multi-year deal to run Perplexity training and inference on GB300 NVL72 clusters — emmanuelvivier · 2026-09-23
- Cost of intelligence falling 50% per quarter, 4x faster than DNA sequencing, says Epoch AI — Terminator857 · 2026-09-23
- DiffusionGemma-Jev lands in vLLM: single-step structured answers with confidence scores — bodonoghue85 · 2026-09-23