3 ms·
This is similar to my experience with (8-bit quantized, non-MOE, 26b) Qwen locally on my computer. It’s really good for small tasks, but the first time I tried
by odo1242 3mo ago
This is similar to my experience with (8-bit quantized, non-MOE, 26b) Qwen locally on my computer. It’s really good for small tasks, but the first time I tried to do a major task with it it straight up forgot what agent harness it was in and started using the wrong format for tool calls lol
(If you’re curious, it was running in Pi, but somehow convinced itself it was running in Claude instead and started trying to call Claude tools that didn’t exist)
- aka-rider 3mo agoModel+harness combination means a lot. That's why all major labs are making their own. All models have quircks harnesses know about "you are reading the same file 3rd time you are in the loop, step back" I tried all frontier Chinese models, and Qwen is the one running the best in ClaudeCode, my personal theory, it's because Qwen was distilled from Opus.