GPQA DiamondVerified
84.9%
GPQA Diamond (AA run)
Models / Xiaomi
MiMo-V2.5
Xiaomi · Released 2026-04-22
Xiaomi retired this API on 21 Oct 2026 at 10:00 Beijing time. It is not a current paid endpoint. MiMo-V2.6-Flash replaces it. Benchmark rows stay on the ledger.
Retired Xiaomi API id. Do not budget a new integration against it. The replacement is MiMo-V2.6-Flash.
Composite58
Context128K
Input / 1M—
Output / 1M—
Knowledge cutoffopen
Output speed36 t/s
Statussolid
Family profile
Best published score in each covered benchmark family.
67/100 avg
- Reasoning
- 85
- Coding
- 44
- Agentic
- 64
- Long context
- 73
4 tested benchmark families
Benchmark placements
Per-benchmark positions
Where MiMo-V2.5 places on each public benchmark source that publishes a row. Rank counts every model with a latest row in the same test — not a universal quality score.
| Benchmark | Family | Rank | Score | Source | Date |
|---|---|---|---|---|---|
| Terminal-Bench | Agentic | #49/ 167 | 64 | Artificial Analysis | 2026-09-28 |
| SciCode | Coding | #70/ 97 | 44 | Artificial Analysis | 2026-09-28 |
| GPQA Diamond | Reasoning | #98/ 184 | 85 | Artificial Analysis | 2026-09-28 |
| Humanity's Last Exam | Reasoning | #103/ 186 | 27 | Artificial Analysis | 2026-09-28 |
| AA-LCR | Long context | #107/ 184 | 73 | Artificial Analysis | 2026-09-28 |
| AA output speed | Performance | #140/ 196 | Artificial Analysis | 2026-10-12 |
Family context
Strongest rows in each family
The three highest-scoring models with pages in each capability family. Where MiMo-V2.5 shows up, it's highlighted.
ReasoningMiMo-V2.5 · 85
- 196
- 295
- 395
Math
- 152
- 252
- 350
CodingMiMo-V2.5 · 44
- 174
- 273
- 370
AgenticMiMo-V2.5 · 64
- 191
- 289
- 389
Long contextMiMo-V2.5 · 73
- 189
- 286
- 385
Newest receipts
6 recentLast 120 days
Benchmark rows added to the public ledger for MiMo-V2.5 in the last 120 days. Older rows live in the full table below.
| Benchmark | Family | Score | Source | Days ago |
|---|---|---|---|---|
| AA output speed | Performance | 36 t/s | Artificial Analysis | 18 |
| AA-LCR | Long context | 73% | Artificial Analysis | 32 |
| GPQA Diamond | Reasoning | 84.9% | Artificial Analysis | 32 |
| Humanity's Last Exam | Reasoning | 27.2% | Artificial Analysis | 32 |
| SciCode | Coding | 43.9% | Artificial Analysis | 32 |
| Terminal-Bench | Agentic | 63.67% | Artificial Analysis | 32 |
Family coverage
Best score per family
A high composite that hides a weak family is a trap. These bars surface the families where this model hasn't been publicly tested, and where it leads.
Price vs. performance
Where MiMo-V2.5 sits on the scatter
Hover any dot for name, score, and input price. Keyboard: tab through the top twelve, or use the ranking below.
AnthropicOpenAIGoogleAlibabaMetaZhipuxAIDeepSeekMiniMaxXiaomiMoonshotOtherMeituanMistralNVIDIACohere
Full benchmark ledger
Every catalog benchmark for MiMo-V2.5. Scores link to the original source; gaps mean no public row exists yet.
6 of 13 catalog benchmarks have a sourced row for MiMo-V2.5.
Reasoning
Humanity's Last ExamVerified
27.2%
HLE (AA run)
Coding
SciCodeVerified
43.9%
SciCode (AA run)
Agentic
Terminal-BenchVerified
63.67%
Terminal-Bench v2.1 (AA run)
Long context
AA-LCRVerified
73%
AA-LCR (AA run)
Performance
AA output speedVerified
36 t/s
Median output tokens/s (1k prompt, default provider)
Changelog
Version history
- Xiaomi retired the V2.5 API id at 10:00 Beijing time. V2.6 Flash replaces it. The old AA figure stays in the pricing note.
Sources
Where the numbers come from
- Artificial Analysis models index· checked 2026-07-26
- Xiaomi MiMo pay-as-you-go· checked 2026-10-21
- Xiaomi MiMo model deprecation· checked 2026-10-21
Questions this model page answers
Every answer below is assembled from the dated fields on this page. Nothing is written separately for search.
What is MiMo-V2.5?
MiMo-V2.5 is a Xiaomi model: Xiaomi retired this API on 21 Oct 2026 at 10:00 Beijing time. It is not a current paid endpoint. MiMo-V2.6-Flash replaces it. Benchmark rows stay on the ledger.
What is MiMo-V2.5 for?
Retired Xiaomi API id. Do not budget a new integration against it. The replacement is MiMo-V2.6-Flash.
What does MiMo-V2.5 cost?
No current Xiaomi token list, checked 21 Oct 2026. Artificial Analysis listed $0.14 input and $0.28 output per 1M tokens on 26 Jul 2026. That figure is not payable.
How large is the MiMo-V2.5 context window?
MiMo-V2.5 publishes a 128K token context window.
What is the VerdictPal status of MiMo-V2.5?
MiMo-V2.5 carries Solid status on the model atlas.
The Pack · Editorial newsletter
Know when the rules move.
One short email when a university changes its AI rule for theses, or a research tool changes its pricing or privacy terms. No tracking, no third-party analytics. Unsubscribe in one click.