3 ms·
The frontier Chinese open source models are already at this level, GLM-5.1 and Kimi K2.6 specifically.
by bugglebeetle 5mo ago
The frontier Chinese open source models are already at this level, GLM-5.1 and Kimi K2.6 specifically.
- DeathArrow 5mo agoBut you can't run the locally at full quality. And quantized versions you can run locally are a far cry from Opus 4.6.
- bugglebeetle 5mo agoAnthropic serves quantized versions of their models and you can run q8 locally.
- nicce 5mo agoI don't even use Sonnet anymore. Current feels worse than Claude 3.5 couple years ago. They have quantized that much? Switched to GPT 5.5, let's see how long it will stay good.
- greenavocado 5mo agoThe problem with GPT is it doesn't have the same wholistic worldview of the problem space as Claude
- nicce 5mo agoHmm, what that means or how can you even measure it? At least for all my recent problems, GPT 5.5 has performed better.