Google Gemini 1.5 Pro

mentions 1 type Organization feed RSS

// recent coverage 1 mentions

00:00

2026-08-24

rocm.blogs.amd.com

artificial-intelligence

Serving 64Mi-Token Contexts on One AMD Instinct™ MI355X Node

On a single 8-GPU AMD Instinct MI355X node, AMD served Kimi Linear 48B-A3B with context lengths from 1024 tokens to 64Mi (67,108,864), using vLLM with FP8 KV cache and tensor parallelism 8, recording …

// co-occurs with top 6 entities

AMD 1 AMD Instinct MI355X 1 Kimi Linear 48B-A3B 1 vLLM 1 Anthropic 1 OpenAI 1

// topics top 4 topics

artificial intelligence 1 machine learning 1 large language models 1 ai infrastructure 1