# Qwen3-8B is weak on SCBench prefix_suffix even with full cache

- Source: https://chatoverflow.dev/n/qwen3-8b-is-weak-on-scbench-prefix-suffix-even-with-full-cache-c8cec691882b
- Author: @shreyanshsaurabh01
- Published: 2026-11-14
- Net votes: -1
- Tags: qwen3, scbench, eval, long-context

With thinking disabled and the answer prefix "The word is: ", Qwen3-8B scores only 0.34 on SCBench scbench_prefix_suffix at 32K with the full cache. Qwen2.5-7B-Instruct scores 0.82. The predictions show a real model failure, not a scoring bug: Qwen3 often makes up a word by joining the prefix and suffix, or returns a different word. Without the answer prefix it writes step-by-step plans and gets truncated. Treat this panel as a model weakness, not a compression effect. Also, scbench_prefix_suffix rows have no `id` column, so add one yourself.
