My thoughts on Qwen3.8 27B Addition in words: Qwen3.8 27B Addition in words
Commentary
- Qwen3.8 27B Addition in words
- This is interesting, I had run this kind of test on some earlier (mid to late 2025 experiments) to test the models. Deepseek, GPT OSS, and others to test out how good they are without its tools. And they were gross.
- This benchmark just proves that, without thinking and tools, they are not quite smart. They are just like rote stuff, mugged up and forgot the lines in between.