Author: Kota Iizuka

Category:

All AI Engineering Edge Robotics Event Reports GPU HPC Networking Systems Development

Day 1 Deployment of DeepSeek-V4-Pro-0813: 17,000 tok/s on Real Traffic, and the Top of a Three-Way Race [B300 x8 Benchmark]

Aug 17, 2026 |

AI Engineering

Day 1 Deployment of DeepSeek-V4-Pro-0813: 17,000 tok/s on Real Traffic, and the Top of a Three-Way Race [B300 x8 Benchmark]

Kota Iizuka

Day 1 with Qwen3.8-2.4T-A95B: Twice Kimi-K3’s Throughput on One 8x B300 Node

Aug 13, 2026 |

AI Engineering

Day 1 with Qwen3.8-2.4T-A95B: Twice Kimi-K3’s Throughput on One 8x B300 Node

Kota Iizuka

Accelerating Inference of gpt-oss-120b with EAGLE-3

Mar 9, 2026 |

AI Engineering

Accelerating Inference of gpt-oss-120b with EAGLE-3

Kota Iizuka

Running Llama 4 Scout on a Single NVIDIA H100 using INT4 Quantization

Jun 26, 2025 |

AI Engineering

Running Llama 4 Scout on a Single NVIDIA H100 using INT4 Quantization

Kota Iizuka

Llama 4 Scout Fine-tuning and Performance Engineering

Jun 23, 2025 |

AI Engineering

Llama 4 Scout Fine-tuning and Performance Engineering

Kota Iizuka

Verifying the Optimal Context Length for Deploying Llama 4 with vLLM

Jun 16, 2025 |

GPU

Verifying the Optimal Context Length for Deploying Llama 4 with vLLM

Kota Iizuka

Thoroughly Testing the New LLM “Gemma 3 (27B)” in a Local Environment: How Does It Perform on Business Tasks?

Jun 5, 2025 |

AI Engineering

Thoroughly Testing the New LLM “Gemma 3 (27B)” in a Local Environment: How Does It Perform on Business Tasks?

Kota Iizuka

Running Llama 4 On-Premise: AI Performance Optimization & Cost Reduction Trial

Jun 2, 2025 |

AI Engineering

Running Llama 4 On-Premise: AI Performance Optimization & Cost Reduction Trial

Kota Iizuka

Using llama.cpp to run Llama 4 Scout on RTX 6000 Ada

May 21, 2025 |

AI Engineering

Using llama.cpp to run Llama 4 Scout on RTX 6000 Ada

Kota Iizuka