I love K2.7. I've used it very heavily for a few weeks. They only reason I don't use it exclusively is that I have my GPT subscription and hence no marginal cost of using GPT Kimi seems better than other models at noticing opportunities for reusing code When I've done work in Kimi, I like asking G...

Kimi 2.7 is okayish compared to GLM 5.2. GLM seems to be better. Don’t see the need to switch to Kimi. For tasks that don’t require GLM’s expertise, I use DeepSeek-V4-Pro which is much much cheaper and would be comparable to Kimi K2.7 (benchmarks are lower but cannot beat on costs). My cur...

Dirt cheap and does the job well (at least with my harness and config).

Terribly but it depends on what you are building

I use Opus 4.8 daily. Where can I run kimi or glm?

I do not have direct experience with Kimi Code K2.7 because running it locally requires insane resources. Even at INT4 quantization it demands nearly 580 GB of VRAM which means an enterprise 8x H100 node. I prefer hosting my LLMs locally on manageable hardware which is why I stick with Qwen 3.6 cu...