#DeepSeek

AIHu Han, former head of Tencent Hunyuan's visual large model algorithm center, is raising tens of millions of dollars for a multimodal startup at a target valuation of several hundred million dollars, with Yuanshi Capital as financial advisor. Investors say Hu has spoken with several firms over the past month and has suggested his model could eventually be sold to large technology companies such as DeepSeek.

AIInferact, working with DeepSeek, NVIDIA, and SemiAnalysis alongside the vLLM community, says joint work across models, custom kernels, and engine serving cuts time to first token (TTFT) by nearly 70% at ~100K throughput. vLLM is the open-source inference engine, and Inferact optimizes it for enterprise production deployments.

AITogether with DeepSeek's model and kernels and NVIDIA's collaboration, vLLM reports that time to first token (TTFT) drops nearly 70% at roughly 100K throughput. The post credits Inferact and the vLLM community for the work and thanks SemiAnalysis for AgentX.

AIvLLM has integrated DeepSeek's MegaAttention, which uses NVFP4 KV cache and is 45% smaller, alongside Mega-mHC, Mega-Gate, and DeepSelect kernels. Its own fusions include a CuTe-DSL fused WO-A that lowers ITL by up to about 6–7%, and sparse MQA logits that run 14–23× faster per layer at 512K context.

AICline says it is pausing its free DeepSeek-V4.1-Flash promotion because of abnormally high abuse. The company says it is actively investigating and working to mitigate the situation.