Skip to content
CheapestInference

MiMo v2.5 API — retired (upgraded to MiMo V2.6 Flash)

MiMo v2.5 was Xiaomi’s fast, efficient model. It was served in the Core Pool until September 23, 2026, when the pool’s MiMo slot was upgraded in place to MiMo V2.6 Flash — a new generation at the same size class, with image input and on-demand reasoning, at the same flat price. Core subscriptions were not interrupted and nothing changed on existing keys.

What this means

Section titled “What this means”
  • Requests with model: "mimo-v2.5" are served by MiMo V2.6 Flash until October 23, 2026. After that date the id returns an invalid-model error.
  • Core Pool subscriptions are unchanged: DeepSeek V4.1 Flash and MiMo V2.6 Flash remain served with no token caps.
  • GET /v1/models is always the authoritative live list.

Migrate in one line

Section titled “Migrate in one line”

Change the model id — the endpoint, key, and subscription stay the same:

from openai import OpenAI


client = OpenAI(
    base_url="https://api.cheapestinference.com/v1",
    api_key="sk-..."  # your subscriber key
)


response = client.chat.completions.create(
    model="mimo-v2.6-flash",  # was "mimo-v2.5"
    messages=[{"role": "user", "content": "Classify these support tickets..."}],
)
Terminal window
curl https://api.cheapestinference.com/v1/chat/completions \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model": "mimo-v2.6-flash", "messages": [{"role": "user", "content": "Hello"}]}'

Coding agents: set the model to mimo-v2.6-flash in your tool’s config — see Claude Code, Cline / Roo Code and OpenCode, Roo & Aider.

MiMo v2.5 in brief (historical)

Section titled “MiMo v2.5 in brief (historical)”
Model IDmimo-v2.5 (retired — use mimo-v2.6-flash)
ServedCore Pool, until 2026-09-23
Per-token price elsewhere$0.14 in / $0.28 out per 1M — reference only
SuccessorMiMo V2.6 Flash — 309B MoE / 15B active, image input, MIT weights, same list price

Common questions

Section titled “Common questions”

Is there still a MiMo v2.5 API? Not as a separate model. Requests for mimo-v2.5 are served by MiMo V2.6 Flash until October 23, 2026; use mimo-v2.6-flash going forward.

Do I lose anything with the upgrade? No — MiMo V2.6 Flash is a newer generation at the same price and the same size class, with image input on the same endpoint. The upgrade was applied to every Core subscription automatically.

Do I need a new subscription or key? No. The same Core Pool subscription and API key serve DeepSeek V4.1 Flash and MiMo V2.6 Flash — just change the model id.

Section titled “Related”