BlogLLM Benchmark Scores as Published, Side by Side
MiMo-V2.5-Pro Base: Benchmark Scores as Published
Every result the catalog holds for MiMo-V2.5-Pro Base: 20 tests in 5 areas, from 1 page. All were printed by Xiaomi itself. Each number is as printed; nothing is ranked or rescored.
Developer XiaomiReleased 2026-04-27Context window 256KInput TextOutput TextWeights OpenData version 2026103001, as of 2026-10-30
Agentic Coding
| Test | Value | Printed by | Setup as printed |
|---|---|---|---|
| SWE-bench (agentless, split not stated) | 35.7% | Xiaomi, Xiaomi MiMo-V2.5-Pro model cardchecked 2026-10-15 | Harness: 3-shot |
Science and Math
| Test | Value | Printed by | Setup as printed |
|---|---|---|---|
| GPQA Diamond | 66.7% | Xiaomi, Xiaomi MiMo-V2.5-Pro model cardchecked 2026-10-15 | Harness: 5-shot |
| MATH | 86.2% | Xiaomi, Xiaomi MiMo-V2.5-Pro model cardchecked 2026-10-15 | Harness: 4-shot |
| GSM8K | 99.6% | Xiaomi, Xiaomi MiMo-V2.5-Pro model cardchecked 2026-10-15 | Harness: 8-shot |
| AIME 2024 and 2025 | 37.3% | Xiaomi, Xiaomi MiMo-V2.5-Pro model cardchecked 2026-10-15 | Harness: 2-shot |
General Knowledge and Reasoning
| Test | Value | Printed by | Setup as printed |
|---|---|---|---|
| MMLU-Pro | 68.5% | Xiaomi, Xiaomi MiMo-V2.5-Pro model cardchecked 2026-10-15 | Harness: 5-shot |
| MMLU-Redux | 92.8% | Xiaomi, Xiaomi MiMo-V2.5-Pro model cardchecked 2026-10-15 | Harness: 5-shot |
| C-Eval | 91.5% | Xiaomi, Xiaomi MiMo-V2.5-Pro model cardchecked 2026-10-15 | Harness: 5-shot |
| MMLU | 89.4% | Xiaomi, Xiaomi MiMo-V2.5-Pro model cardchecked 2026-10-15 | Harness: 5-shot |
| DROP (F1) | 86.3% | Xiaomi, Xiaomi MiMo-V2.5-Pro model cardchecked 2026-10-15 | Harness: 3-shot |
| BBH | 88.4% | Xiaomi, Xiaomi MiMo-V2.5-Pro model cardchecked 2026-10-15 | Harness: 3-shot |
| CMMLU | 90.2% | Xiaomi, Xiaomi MiMo-V2.5-Pro model cardchecked 2026-10-15 | Harness: 5-shot |
| HellaSwag | 89.8% | Xiaomi, Xiaomi MiMo-V2.5-Pro model cardchecked 2026-10-15 | Harness: 10-shot |
| TriviaQA | 81.3% | Xiaomi, Xiaomi MiMo-V2.5-Pro model cardchecked 2026-10-15 | Harness: 5-shot |
| WinoGrande | 85.6% | Xiaomi, Xiaomi MiMo-V2.5-Pro model cardchecked 2026-10-15 | Harness: 5-shot |
| ARC-Challenge | 97.2% | Xiaomi, Xiaomi MiMo-V2.5-Pro model cardchecked 2026-10-15 | Harness: 25-shot |
Code Generation
| Test | Value | Printed by | Setup as printed |
|---|---|---|---|
| LiveCodeBench v6 | 39.6% | Xiaomi, Xiaomi MiMo-V2.5-Pro model cardchecked 2026-10-15 | Harness: 1-shot |
| HumanEval+ | 75.6% | Xiaomi, Xiaomi MiMo-V2.5-Pro model cardchecked 2026-10-15 | Harness: 1-shot |
| MBPP+ | 74.1% | Xiaomi, Xiaomi MiMo-V2.5-Pro model cardchecked 2026-10-15 | Harness: 3-shot |
Multilingual
| Test | Value | Printed by | Setup as printed |
|---|---|---|---|
| Global-MMLU | 83.6% | Xiaomi, Xiaomi MiMo-V2.5-Pro model cardchecked 2026-10-15 | Harness: 5-shot |
More from Xiaomi
Every number here is read from the same catalog file as the benchmark table, data version 2026103001. Where one page printed the model under several setups, each value is listed. The catalog is public in armanobosyan/sugra-llm-benchmarks; a wrong number can be reported from the table, with Report an error under the value, or as an issue in that repository.