#95 MiMo-V2.5
none- Cost
- $0.007
- Time
- 267.4s
- Tokens
- 25,283 tok
Xiaomi
Release: 2026-04-22
Tested on: 2026-04-22 21:39
xiaomi/mimo-v2.5::none
310B total (15B active)MoEOpen source
(medium)
(none)
Summary
MiMo-V2.5 scores 5.1 on AI BENCHY and ranks #95. It has N/A reliability, a 27.8% pass rate, $0.019 total cost, and 1.05s average response time.
What makes MiMo-V2.5 unique: It stands out most in Coding, where it ranks #1, while Anti-AI Tricks is its weakest area at #17. Its total benchmark cost is unusually low for its score range. It is notably fast compared with similar models.
Model facts
Researched on 2026-08-12
- Parameters
- 310B total (15B active)
- Architecture
- MoE
- Availability
- Open source
- License
- MIT
5.1
Consistency
10.0
N/A
$0.019
Total Output Tokens
2,177
Total Input Tokens
0
Input Price
$0.400 / 1M
Output Price
$2.000 / 1M
Flaky tests
0
Flaky tests had mixed outcomes across runs (at least one pass and one fail).
Hamster playing table tennis
Prompt: Create a detailed SVG illustration of a hamster playing table tennis.
Run history
| Tested on | Score | Reliability | Tests Correct | Total Cost | Compare |
|---|---|---|---|---|---|
| 2026-08-14 02:25 Re-test | 5.1 | 10.0 | $0.025 ↓ | Compare | |
| 2026-07-16 22:42 New test added | 5.1 | 10.0 | $0.025 ↓ | Compare | |
| 2026-06-04 13:48 New test added | 4.9 | 10.0 | $0.007 ↓ | Compare | |
| 2026-05-22 12:51 New test added | 4.8 | 10.0 | $0.021 | Compare | |
| 2026-05-08 15:33 Suite changed | 4.9 | 10.0 | $0.019 | Compare | |
| 2026-04-22 21:39 First recorded run | 5.1 | N/A | $0.019 | Current run |
Run comparison
| Run | Benchmark coverage | Score | Consistency | Reliability | Tests Correct | Flaky tests | Total Output Tokens | Total Input Tokens | Total Cost | Response Time (avg) |
|---|---|---|---|---|---|---|---|---|---|---|
| 2026-04-22 21:39 · First recorded run | 54/54 attempts | 5.1 | 10.0 | N/A | 5/18 | 0 | 2,177 | 0 | $0.019 | 1.05s |
| 2026-06-04 13:48 · New test added | 63/63 attempts | 4.9 | 9.6 | 10.0 | 5/21 | 1 | 2,267 | 41,985 | $0.007 | 2.20s |
| Difference | — | +0.2 | +0.4 | 0 | -1 | -90 | -41985 | +$0.012 | -1148ms |
Benchmark coverage differs: 54/54 attempts (Target: 3 repeats per test) versus 63/63 attempts (Target: 3 repeats per test). Totals and repeat-sensitive metrics are not directly comparable.
These two runs used different benchmark suites, so the deltas reflect both model changes and suite changes.
Charts
Choose the first model, then click a second model to open a side-by-side page.
Top Models by Score
Score vs Total Cost
Response Time (avg)
Score vs Response Time (avg)
Total Output Tokens
Score vs Total Output Tokens
Quick Compare
MiMo-V2.5nonevsQwen3 Coder NextnoneMiMo-V2.5nonevsQwen3.5-9BnoneMiMo-V2.5nonevsGranite 4.2 8BlowMiMo-V2.5nonevsNorth Mini CodenoneFree AvailableMiMo-V2.5nonevsMistral Small 4mediumMiMo-V2.5nonevsMiniMax M2.7mediumMiMo-V2.5nonevsMistral Small 4noneMiMo-V2.5nonevsMercury 2.5 PreviewlowMiMo-V2.5nonevsNemotron 3.5 LightningmediumFree AvailableMiMo-V2.5nonevsLaguna S 2.1lowFree Available
Category Breakdown
| Category | Score | Consistency | Tests Correct |
|---|---|---|---|
| Anti-AI Tricks | 4.8 | 10.0 | |
| Coding | 10.0 | 10.0 | |
| Combined | 3.0 | 10.0 | |
| Data parsing and extraction | 6.5 | 10.0 | |
| Domain specific | 3.0 | 10.0 | |
| General Intelligence | 4.6 | 10.0 | |
| Instructions following | 6.5 | 10.0 | |
| Puzzle Solving | 3.4 | 10.0 | |
| Tool Calling | 10.0 | 10.0 |