RTX Spark vs M5 Max: local Qwen 3.8 27B benchmark shows NVIDIA wins on compute
draginol · x · 2026-10-19
The author compares flagship laptops running local models. Screenshots show NVIDIA RTX Spark N1X (6144-core Blackwell GPU, 64GB RAM) hitting 1,636 tok/s prefill with 971ms first token, while Apple M5 Max (128GB unified memory) running qwen3.8:27b-mlx reaches 577 tok/s prefill and 43.9 tok/s decode with a 1.65s first token. His takeaway: a tricky result — M5 Max has higher memory bandwidth, but its compute isn't close to NVIDIA's.
Related event: RTX Spark vs M5 Max: Local Qwen 3.8 27B Inference Benchmark(2 posts)→
More from Infra
- CoreWeave plunges 25% in a week as credit default swaps hit distress levels — MartinGTobias · 2026-10-19
- vLLM cuts container startup time 14-18% with one uv bytecode flag — vllm_project · 2026-10-19
- Why neoclouds get acquired by outsiders, never by each other — AccBalanced · 2026-10-19
- Jev Router cuts coding agent costs 67-85% by routing tasks to cheap open models — HowDevelop · 2026-10-19
- Baidu's Virtual Head Attention cuts KV cache 4-7x with zero custom kernels — burny_tech · 2026-10-19
- Server memory spot growth slows to +0.6% as storage stocks fall 5-7% in a day — tengyanAI · 2026-10-19