For 10,000 requests of 2,000 tokens in and 500 out, MiniMax M2.5 costs 23% less than MiniMax M3 on Zurelay: $3.95 against $5.15. Both are below their labs' list prices: 68% for MiniMax M2.5 and 57% for MiniMax M3.
Side by side
| MiniMax M2.5 | MiniMax M3 | |
|---|---|---|
| Lab | MiniMax | MiniMax |
| Released | Feb 12, 2026 | Jun 1, 2026 |
| Context window | 205K tokens | 1M tokens |
| Max output | 128K tokens | — |
| Reads | Text | Text, Images |
| Reasoning levels | Always on, no levels | none, high |
| Price per 1M, in / out | $0.10 / $0.39 | $0.13 / $0.51 |
| Lab list price | $0.30 / $1.20 | $0.30 / $1.20 |
| Cached input per 1M | $0.01 | $0.026 |
| 10K typical requests | $3.95 | $5.15 |
Prices in US dollars per 1M tokens on Zurelay, live from the catalog. A typical request is 2,000 tokens in and 500 out. Output limits are each lab's.
What they're for
MiniMax M2.5 is MiniMax's open-weight model for coding, agentic tool use and search, and office work, released on February 12, 2026. The Zurelay MiniMax M2.5 API serves it through one OpenAI-compatible endpoint with a 204,800-token context window and up to 131,072 output tokens per answer, at 68% below MiniMax's list price.
MiniMax M3 is MiniMax's natively multimodal model for agentic coding, tool use and long context, released on June 1, 2026, with about 428B total and 23B active parameters. The Zurelay MiniMax M3 API serves it through one OpenAI-compatible endpoint with a 1M-token context window and image input, at 57% below MiniMax's list price.
Try both
Specs don't say which model handles your work better. Send the same requests to both and compare the answers, the tokens and the bill. On Zurelay that's a change of model name.
To price a whole month, use the LLM cost calculator.
from openai import OpenAIclient = OpenAI(base_url="https://api.zurelay.com/v1", api_key="YOUR_ZURELAY_KEY")prompt = [{"role": "user", "content": "Summarize this contract: ..."}]# Same key, same code: only the model name changesfor model in ["minimax-m2.5", "minimax-m3"]: reply = client.chat.completions.create(model=model, messages=prompt) print(model, reply.usage.total_tokens, reply.choices[0].message.content[:200])For 10,000 requests of 2,000 tokens in and 500 out, MiniMax M2.5 costs 23% less than MiniMax M3 on Zurelay: $3.95 against $5.15. Per 1M tokens, MiniMax M2.5 is $0.10 input and $0.39 output, and MiniMax M3 is $0.13 and $0.51.
MiniMax M3, at 1M tokens, against 205K for MiniMax M2.5.
Yes. One Zurelay key calls both. With an OpenAI SDK, point it at https://api.zurelay.com/v1 and change only the model name, minimax-m2.5 or minimax-m3, to switch.
Yes. Each model is served over several routes run by independent sellers, and every route is tested before it serves requests and every day after, with identity, dated-knowledge and reasoning checks that catch an older or different model sold under the name you asked for.
Create a key in seconds, point the SDK you already use at Zurelay, and every request costs up to 90% less from the first token.