Data points: 109
Model compare
The readout for DeepSeek V4 Pro and MiMo-V2.5, before the detailed comparison sheet.
Weighted outcome: MiMo-V2.5. Benchmark capability categories carry 80%, while price, API performance, and availability carry 20%.
Decision read
MiMo-V2.5
MiMo-V2.5 has the higher weighted result; Model A / B score 34.7 to 65.3.
Evidence depth
109 data points
Includes 19 benchmark rows, 1 audit samples, and 6 provider examples.
Selection signal
Start with MiMo-V2.5
The charts below split 26 high-signal samples across speed, scores, and audit health.
Switch either side of this report to compare another model with the same LMSpeed data pipeline.
Select a different model to open a new comparison URL.
This report only uses LMSpeed data for DeepSeek V4 Pro and MiMo-V2.5: pricing, speed aggregates, third-party benchmark scores, and shared provider samples.
| Model compare | DeepSeek V4 Pro | MiMo-V2.5 |
|---|---|---|
| Overall leader | Contender | Leading |
| Weighted overall score | 34.7 pts | 65.3 pts |
| Benchmark category leads | 1 categories | 2 categories |
| Operational advantages | Free providers, Provider coverage | Cheapest input price, Average speed, First-token latency |
| Context window | 1.0M tokens | 1.1M tokens |
| Max output | 384K tokens | 131.1K tokens |
| Modalities | Input Text Output Text | Input Text |
The overall result weights benchmark capability categories at 80% and price, API speed/latency, and availability at 20%. Recent test volume does not affect the winner, and missing benchmark categories are excluded.
| Model compare | DeepSeek V4 Pro | MiMo-V2.5 |
|---|---|---|
| Developer | DeepSeek | Xiaomi |
| Released | Apr 2026 | Apr 2026 |
| Parameters | No data | No data |
| Tokenizer | DeepSeek | Other |
| Knowledge cutoff | No data | No data |
| OpenRouter ID | deepseek/deepseek-v4-pro | xiaomi/mimo-v2.5 |
| References | No data | No data |
This report only uses LMSpeed data for DeepSeek V4 Pro and MiMo-V2.5: pricing, speed aggregates, third-party benchmark scores, and shared provider samples.
DeepSeek V4 Pro
DeepSeek V4 Pro is stronger in benchmark categories (Agents) and operational dimensions (Free providers, Provider coverage).
MiMo-V2.5
MiMo-V2.5 is stronger in benchmark categories (Coding, Reasoning) and operational dimensions (Cheapest input price, Average speed, First-token latency).
Third-party benchmark profile synced into LMSpeed; only metrics available for both models are shown.
Compare benchmark category scores on a 0-100 scale. Select a category to inspect the gap.
Avg. score
DeepSeek V4 Pro
47.4
Avg. score
MiMo-V2.5
51.8
DeepSeek V4 Pro leads by 0.6
MiMo-V2.5 leads by 2.9
MiMo-V2.5 leads by 2.5
DeepSeek V4 Pro
DeepSeek V4 Pro
No data
MiMo-V2.5
DeepSeek V4 Pro
Metric-level scores with benchmark source, rank depth, confidence, error, and evaluation date where available.
DeepSeek V4 Pro
winner65.0
Rank #25/93 · confidence 2 · eval date 2026-04-24
MiMo-V2.5
55.0
Rank #52/93 · confidence 2 · eval date 2026-04-22
DeepSeek V4 Pro
winner1.22 s
Rank #35/75 · confidence 4
MiMo-V2.5
5.24 s
Rank #61/75 · confidence 4
DeepSeek V4 Pro
55.9 tok/s
Rank #59/75 · confidence 4
MiMo-V2.5
winner77.3 tok/s
Rank #50/75 · confidence 4
DeepSeek V4 Pro
36.84 s
Rank #63/75 · confidence 4
MiMo-V2.5
winner31.10 s
Rank #58/75 · confidence 4
DeepSeek V4 Pro
$0.870/M
Rank #37/174 · confidence 4
MiMo-V2.5
winner$0.280/M
Rank #8/174 · confidence 4
DeepSeek V4 Pro
$0.544/M
Rank #52/174 · confidence 4
MiMo-V2.5
winner$0.175/M
Rank #12/174 · confidence 4
DeepSeek V4 Pro
$0.435/M
Rank #72/174 · confidence 4
MiMo-V2.5
winner$0.140/M
Rank #18/174 · confidence 4
DeepSeek V4 Pro
winner90.5%
Rank #21/200 · confidence 4
MiMo-V2.5
84.9%
Rank #53/200 · confidence 4
DeepSeek V4 Pro
winner35.2%
Rank #31/199 · confidence 4
MiMo-V2.5
27.2%
Rank #54/199 · confidence 4
DeepSeek V4 Pro
winner67.9
Rank #17/31 · confidence 2 · eval date 2026-04-24
MiMo-V2.5
65.8
Rank #20/31 · confidence 2 · eval date 2026-04-22
DeepSeek V4 Pro
55.4
Rank #32/45 · confidence 2 · eval date 2026-04-24
MiMo-V2.5
winner56.1
Rank #29/45 · confidence 2 · eval date 2026-04-22
DeepSeek V4 Pro
winner46.4%
Rank #37/197 · confidence 4
MiMo-V2.5
43.1%
Rank #50/197 · confidence 4
DeepSeek V4 Pro
winner50.4
Rank #43/74 · confidence 2 · eval date 2026-04-24
MiMo-V2.5
37.3
Rank #66/74 · confidence 2 · eval date 2026-04-22
DeepSeek V4 Pro
1259.0
Rank #31/70 · confidence 2 · eval date 2026-04-24
MiMo-V2.5
winner1293.0
Rank #20/70 · confidence 2 · eval date 2026-04-22
DeepSeek V4 Pro
59.8
Rank #14/28 · confidence 2 · eval date 2026-04-24
MiMo-V2.5
winner62.3
Rank #10/28 · confidence 2 · eval date 2026-04-22
DeepSeek V4 Pro
winner17.1
Rank #11/19 · confidence 2 · eval date 2026-04-24
MiMo-V2.5
16.9
Rank #13/19 · confidence 2 · eval date 2026-04-22
DeepSeek V4 Pro
winner67.9
Rank #20/51 · confidence 2 · eval date 2026-04-24
MiMo-V2.5
65.8
Rank #23/51 · confidence 2 · eval date 2026-04-22
DeepSeek V4 Pro
winner67.5
Rank #22/62 · confidence 2 · eval date 2026-04-24
MiMo-V2.5
58.1
Rank #34/62 · confidence 2 · eval date 2026-04-22
DeepSeek V4 Pro
winner50.3
Rank #23/50 · confidence 2 · eval date 2026-04-24
MiMo-V2.5
46.9
Rank #27/50 · confidence 2 · eval date 2026-04-22
Latest completed audits from shared providers, with four safety and integrity score groups plus report links.
| Provider | DeepSeek V4 Pro | MiMo-V2.5 |
|---|---|---|
Winner: DeepSeek V4 Pro | DeepSeek V4 Pro deepseek-v4-pro Audit score 87 788486100 | MiMo-V2.5 mimo-v2.5 |
Speed aggregates and input/output pricing share each provider row for real API selection and migration cost checks.
| Provider | DeepSeek V4 Pro | MiMo-V2.5 |
|---|---|---|
40 tests | DeepSeek V4 Pro speed / latency 53 tok/s / 13840ms input / output No data | MiMo-V2.5 speed / latency 72 tok/s / 3752ms input / output No data |
20 tests | DeepSeek V4 Pro speed / latency |
Weighted outcome: MiMo-V2.5. Benchmark capability categories carry 80%, while price, API performance, and availability carry 20%.
Output
| Features | Text inputText outputTool callingStructured outputsJSON modeReasoning | Text inputImage inputAudio inputText outputTool callingStructured outputsJSON modeReasoning |
|---|
No audit yet
41 tok/s / 9833ms
input / output
No data
MiMo-V2.5 speed / latency 86 tok/s / 2076ms input / output No data |
6345ywz API15 tests | DeepSeek V4 Pro speed / latency 27 tok/s / 51460ms input / output No data | MiMo-V2.5 speed / latency 66 tok/s / 6202ms input / output No data |
|---|
PICO AI10 tests | DeepSeek V4 Pro speed / latency 32 tok/s / 11322ms input / output No data | MiMo-V2.5 speed / latency 97 tok/s / 1664ms input / output No data |
|---|
Xinjianya API10 tests | DeepSeek V4 Pro speed / latency 28 tok/s / 6536ms input / output No data | MiMo-V2.5 speed / latency N/A / N/A input / output No data |
|---|
DeepSeek V4 Pro deepseek-v4-pro speed / latency No data input / output $0.167/M/$0.334/M | MiMo-V2.5 mimo-v2.5 speed / latency No data input / output $0/M/$0/M |
