moonshotai
Kimi K2.7 Code
1T-parameter (32B activated) Kimi model specialised for coding, with 262,144-token context.Modified MITHugging Face
No server in stock right now has the 714 GB this model needs for inference.
License restrictions
Products above 100M monthly active users or USD 20M monthly revenue must prominently display the model name.
GPU memory it needs
From model cards and vendor documentation. Lower precision needs less memory, with slightly lower quality. Estimates are marked.
| What you want to do | Precision | Minimum | Recommended | Setup | Basis |
|---|---|---|---|---|---|
| Run the model | 4-bit | 714 GB | 1128 GB | 8x H200 (vLLM recipe default, single-node) | Published |