August 18, 2026
Qwen3.8-27B on an M4 Pro
Throughput, memory and benchmark scores for Qwen3.8-27B at 4-bit on a 48 GB MacBook Pro, and the generation cap that was setting most of the accuracy numbers.
Read articleTag Archive
3 posts collected under this topic.
Entries connected to LLMs.
August 18, 2026
Throughput, memory and benchmark scores for Qwen3.8-27B at 4-bit on a 48 GB MacBook Pro, and the generation cap that was setting most of the accuracy numbers.
Read articleJuly 15, 2026
The instruction files I use to make Claude Fable 5 and GPT-5.6 Sol delegate execution to cheaper subagent models: the tiering rules, the executor definitions, and the briefing format.
Read articleApril 17, 2026
Anthropic is pitching better long-running coding and vision. Hacker News readership comment about consistency, token burn, and whether Claude still feels reliable from one week to the next.
Read article