Latest

The cost crossover at 6.8K tokens/s/chip reframes the Ascend-versus-Blackwell debate from raw benchmark losses to workload-dependent economics, suggesting DeepSeek could run its API on a sub-US$1B Blackwell cluster while pivoting post-training to Ascend. Practitioners tracking inference infrastructure or Huawei's CANN ecosystem get a rare concrete per-chip price and throughput comparison. The proposed 128-chip Ascend supernode topology has no Nvidia equivalent, which matters for anyone planning large-scale decode deployments.

Teortaxes@teortaxesTex11 sourcesSource

The back-of-envelope math suggests DeepSeek's RL training pipeline is compute-bound at the GPU level, not CPU-bound, which changes how analysts should read its published infrastructure claims. Investors and policy watchers tracking US export-control impact on Chinese labs can use the 3,000-GPU figure as a yardstick for scaling bottlenecks. Compare it with NVIDIA Blackwell allocation tiers announced for US hyperscalers.

Teortaxes@teortaxesTexSource

DeepSeek V4 Flash shows strong adoption as the only open-source model competing with proprietary options in enterprise usage.

官方账号LangChain@LangChainAI4 sourcesSource

有人设想让 V4 Pro 规划、V4 Flash 执行，还招开源开发者内测 DeepSeek Harness，想玩 Agent 的可以关注。

Geek@geekbb4 sourcesSource

AISI实测发现开放模型追封闭模型只差几个月，GLM-5.2和DeepSeek V4-Pro表现接近，差距还在快速缩小，值得关注。

@koltregaskes@koltregaskes5 sourcesSource