Ollama проигнорировал мой шаблон Modelfile. Предварительно зарегистрированное количество токенов это поймало.
DEV · 15.11.2026 · 00:00
В ходе тестирования на локальном сервере Qwen3.8 27B GGUF было обнаружено несоответствие в количестве токенов, что привело к недействительности запуска.
Оригинал
Ollama Ignored My Modelfile TEMPLATE. A Pre-Registered Token Count Caught It.
Originally published on hexisteme notes. On November 8 an overnight run on my local Qwen3.8 27B GGUF finished, and before looking at the score I checked one number per task: how many prompt tokens the server reported beyond the tokens of the message body. The pre-registration said that number should be about 14 on all 10 tasks. It was 52 on all 10. By the rule I had written two days earlier, the run was invalid and got no verdict. The variable I was testing was a prompt change made through the