StarAPI
kimi-k2.8-preview vs kimi-k3
| kimi-k2.8-preview | kimi-k3 | |
|---|---|---|
| Providers | Moonshot AI | Moonshot AI |
| Input $/M | $0.98 | $3.03 |
| Output $/M | $2.01 | $15.15 |
| Output multiplier | ×2.05 | ×5.00 |
| Context window | 1M tokens | 1M tokens |
| Intelligence index | — | 50.2 |
| Coding index | — | 76.2 |
What the numbers mean
kimi-k2.8-preview is roughly 3.1× cheaper on input than kimi-k3. If your workload passes the quality bar on the cheaper model, that gap compounds across every request; route to kimi-k3 deliberately rather than by default.
Tags: kimi-k2.8-preview — Chat, Reasoning, Coding, Vision · kimi-k3 — Chat, Reasoning, Coding, Vision
Code
from openai import OpenAI
client = OpenAI(base_url="https://starseaapi.com/v1", api_key="sk-...")
# kimi-k2.8-preview
client.chat.completions.create(model="kimi-k2.8-preview", messages=[{"role":"user","content":"Hi"}])
# kimi-k3
client.chat.completions.create(model="kimi-k3", messages=[{"role": "user", "content": "Hi"}])