---
id: "kt/midm-k-3.0-pro"
also: ["kt/midm-k-3.0-pro"]
card_usd_per_m: "0.2 in, 0.8 out, at kt"
cheapest_list_usd_per_m: "0.2 in, 0.8 out, at kt on openrouter"
cheapest_effective_usd_per_m: "0.1831 in, 0.7995 out, at kt on openrouter"
context: 261795
max_output: 235615
input: ["text"]
output: ["text"]
released: "2026-11-05"
api: ["/api/v1/chat/completions", "/api/v1/completions", "/api/v1/messages", "/api/v1/responses"]
params: ["frequency_penalty", "include_reasoning", "logprobs", "max_tokens", "presence_penalty", "reasoning", "repetition_penalty", "response_format", "structured_outputs", "temperature", "top_k", "top_logprobs", "top_p"]
endpoints: 1
gateways: "openrouter (1)"
no_training: 1
trains: 0
retains_prompts: 1
---
# KT: Mi:dm K 3.0 Pro

Mi:dm K 3.0 Pro is a Mixture-of-Experts reasoning model from KT, with 120B total and 12B active parameters. It uses a hybrid architecture that mixes Mamba, MoE and attention layers.