StarSeaAPI
k3 vs kimi-k2.7-code
| k3 | kimi-k2.7-code | |
|---|---|---|
| Providers | Moonshot AI | Moonshot AI |
| Input $/M | $3.03 | $0.98 |
| Output $/M | $15.15 | $4.09 |
| Output multiplier | ×5.00 | ×4.15 |
| Context window | 1M tokens | 256K tokens |
| Intelligence index | — | — |
| Coding index | — | 60.8 |
What the numbers mean
kimi-k2.7-code is roughly 3.1× cheaper on input than k3. If your workload passes the quality bar on the cheaper model, that gap compounds across every request; route to k3 deliberately rather than by default.
Tags: k3 — Chat, Reasoning, Coding, Vision · kimi-k2.7-code — Chat, Reasoning, Coding, Vision
Code
from openai import OpenAI
client = OpenAI(base_url="https://starseaapi.com/v1", api_key="sk-...")
# k3
client.chat.completions.create(model="k3", messages=[{"role":"user","content":"Hi"}])
# kimi-k2.7-code
client.chat.completions.create(model="kimi-k2.7-code", messages=[{"role": "user", "content": "Hi"}])