StarAPI
minimax-m2.5 vs minimax-m3
| minimax-m2.5 | minimax-m3 | |
|---|---|---|
| Providers | MiniMax | MiniMax |
| Input $/M | $0.32 | $0.64 |
| Output $/M | $1.27 | $2.55 |
| Output multiplier | ×4.00 | ×4.00 |
| Context window | 204.8K tokens | 1M tokens |
| Intelligence index | — | 35.7 |
| Coding index | — | 58.6 |
What the numbers mean
minimax-m2.5 is roughly 2.0× cheaper on input than minimax-m3. If your workload passes the quality bar on the cheaper model, that gap compounds across every request; route to minimax-m3 deliberately rather than by default.
Tags: minimax-m2.5 — Chat, Reasoning, Coding · minimax-m3 — Chat, Reasoning, Coding, Vision
Code
from openai import OpenAI
client = OpenAI(base_url="https://starseaapi.com/v1", api_key="sk-...")
# minimax-m2.5
client.chat.completions.create(model="minimax-m2.5", messages=[{"role":"user","content":"Hi"}])
# minimax-m3
client.chat.completions.create(model="minimax-m3", messages=[{"role": "user", "content": "Hi"}])