All AI models · LLM benchmarks · Methodology

Xiaomi

MiMo-V2.5-Pro

Stronger MiMo Pro tier for multimodal reasoning and coding-agent execution

Model facts

Context window
1.05M tokens
Maximum output
131K tokens
Cheapest paid input
$0.4 per 1M tokens
Cheapest paid output
$0.8 per 1M tokens
Open weights
yes
Providers
32

Capabilities and modalities

reasoning, tool calling, structured output, attachments, temperature control, open weights, text, audio.

Local hardware estimate

At 4K context: Q4_K_M 627.3 GiB, Q8_0 1103.7 GiB, F16 1997.1 GiB working memory.

Parameters
1023.2 billion
Architecture
mimov2
Model license
mit

Planning estimate adapted from the MIT-licensed whichllm estimator using metadata from Hugging Face; not a vendor minimum requirement.

Artificial Analysis benchmarks

Compare this model on the full LLM benchmark leaderboard.

ConfigurationAAICodingMathOutput tokens/s
default42.960.2not measured37.6
Non-reasoning28.4not measurednot measured32.0

Observed price history

  • 2026-08-24: $0.392 input / $0.783 output per 1M tokens
  • 2026-08-28: $0.4 input / $0.8 output per 1M tokens
  • 2026-09-01: $0.4 input / $0.8 output per 1M tokens

API providers