Model comparison
GPT-4 Turbo vs MiMo-V2.5-Pro
MiMo-V2.5-Pro is the stronger model overall, scoring 45.2 to 30.5 on the Noometry Index.
Last verified . 19 shared benchmarks.
Summary
- They share 19 benchmarks with published results for both. GPT-4 Turbo scores higher in 0 categories and MiMo-V2.5-Pro in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where MiMo-V2.5-Pro leads 40.0 to 9.0.
- The biggest single-benchmark swing is DTBench: 61.6% for GPT-4 Turbo and 84.5% for MiMo-V2.5-Pro.
- MiMo-V2.5-Pro is cheaper at $0.43 / $0.87 per million input/output tokens, against $10 / $30 for GPT-4 Turbo.
- MiMo-V2.5-Pro accepts more context: 1.05M tokens versus 128K.
- MiMo-V2.5-Pro has downloadable open weights; the other is API-only.
Side by side
| GPT-4 Turbo | MiMo-V2.5-Pro | |
|---|---|---|
| Provider | OpenAI | Xiaomi |
| Noometry Index | 30.5 | 45.2 |
| Released | 2023-11-06 | 2026-04-22 |
| Weights | Proprietary | Open |
| Context window | 128K | 1.05M |
| Max output | 4K | 131K |
| Input $ / M tokens | $10 | $0.43 |
| Output $ / M tokens | $30 | $0.87 |
| Results tracked | 36 | 27 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding MiMo-V2.5-Pro leads
GPT-4 Turbo: 33.8 (#249), MiMo-V2.5-Pro: 47.4 (#60)
| Benchmark | GPT-4 Turbo | MiMo-V2.5-Pro |
|---|---|---|
| LMArena Coding | 1268 | 1503 |
| LMArena WebDev | — | 1479 |
| SciCode | — | 50.2% |
| WeirdML | 18% | — |
| BigCodeBench Instruct | 48.2% | — |
| BigCodeBench Complete | 58.2% | — |
| ALE-Bench | — | 899.8 |
| HumanEval+ | 86.6% | — |
| MBPP+ | 73.3% | — |
Agentic & Tool Use Not comparable
GPT-4 Turbo: —, MiMo-V2.5-Pro: —
| Benchmark | GPT-4 Turbo | MiMo-V2.5-Pro |
|---|---|---|
| METR Time Horizons | 36.7% | — |
Reasoning MiMo-V2.5-Pro leads
GPT-4 Turbo: 15.3 (#317), MiMo-V2.5-Pro: 26.8 (#130)
| Benchmark | GPT-4 Turbo | MiMo-V2.5-Pro |
|---|---|---|
| LMArena Hard Prompts | 1251 | 1488 |
| DTBench | 61.6% | 84.5% |
| LMCA | 9.8% | 29.5% |
| SimpleBench | 25.1% | — |
| NYT Connections (extended) | — | 34.4% |
| CritPt | — | 4% |
| Chess Puzzles | 6% | — |
| Epoch Capabilities Index | 127.25 | — |
| ForecastBench | 59.4 | — |
Math MiMo-V2.5-Pro leads
GPT-4 Turbo: 9.0 (#322), MiMo-V2.5-Pro: 40.0 (#96)
| Benchmark | GPT-4 Turbo | MiMo-V2.5-Pro |
|---|---|---|
| LMArena Math | 1272 | 1481 |
| FrontierMath (Tiers 1-3) | 0.7% | — |
| OTIS Mock AIME 2024-2025 | 6.7% | — |
| ProofBench | — | 22% |
| MATH Level 5 | 46.7% | — |
Knowledge MiMo-V2.5-Pro leads
GPT-4 Turbo: 24.3 (#268), MiMo-V2.5-Pro: 42.2 (#98)
| Benchmark | GPT-4 Turbo | MiMo-V2.5-Pro |
|---|---|---|
| LMArena Expert | 1223 | 1503 |
| GPQA Diamond | 46.6% | — |
| Confabulations | 28.4% | — |
| MMLU | 81.3% | — |
Multimodal Not comparable
GPT-4 Turbo: 30.6 (#110), MiMo-V2.5-Pro: —
| Benchmark | GPT-4 Turbo | MiMo-V2.5-Pro |
|---|---|---|
| LMArena Vision | 1090 | — |
Multilingual MiMo-V2.5-Pro leads
GPT-4 Turbo: 40.5 (#216), MiMo-V2.5-Pro: 55.1 (#34)
| Benchmark | GPT-4 Turbo | MiMo-V2.5-Pro |
|---|---|---|
| LMArena Non-English | 1245 | 1449 |
| LMArena Chinese | 1242 | 1507 |
| LMArena French | 1276 | 1488 |
| LMArena German | 1259 | 1458 |
| LMArena Japanese | 1194 | 1412 |
| LMArena Korean | 1187 | 1437 |
| LMArena Russian | 1259 | 1450 |
| LMArena Spanish | 1260 | 1471 |
Instruction Following MiMo-V2.5-Pro leads
GPT-4 Turbo: 65.8 (#216), MiMo-V2.5-Pro: 77.5 (#21)
| Benchmark | GPT-4 Turbo | MiMo-V2.5-Pro |
|---|---|---|
| LMArena Instruction Following | 1249 | 1477 |
Long Context MiMo-V2.5-Pro leads
GPT-4 Turbo: 38.0 (#206), MiMo-V2.5-Pro: 45.4 (#37)
| Benchmark | GPT-4 Turbo | MiMo-V2.5-Pro |
|---|---|---|
| LMArena Longer Query | 1254 | 1483 |
Writing & Preference MiMo-V2.5-Pro leads
GPT-4 Turbo: 47.7 (#206), MiMo-V2.5-Pro: 65.3 (#49)
| Benchmark | GPT-4 Turbo | MiMo-V2.5-Pro |
|---|---|---|
| LMArena Text | 1272 | 1465 |
| LMArena Creative Writing | 1269 | 1440 |
| LMArena Multi-Turn | 1267 | 1477 |
| EQ-Bench Creative Writing | — | 1493 |
| EQ-Bench 4 | — | 1208 |
Frequently asked questions
Is GPT-4 Turbo better than MiMo-V2.5-Pro?
MiMo-V2.5-Pro is the stronger model overall, scoring 45.2 to 30.5 on the Noometry Index.
Which is cheaper, GPT-4 Turbo or MiMo-V2.5-Pro?
MiMo-V2.5-Pro is cheaper. It lists at $0.43 per million input tokens and $0.87 per million output tokens; GPT-4 Turbo lists at $10 and $30.
Is GPT-4 Turbo or MiMo-V2.5-Pro better for coding?
MiMo-V2.5-Pro scores higher on coding benchmarks: 47.4 versus 33.8 in the Noometry coding category.
Which has the bigger context window?
MiMo-V2.5-Pro does, with 1.05M tokens against 128K.
How many benchmarks do GPT-4 Turbo and MiMo-V2.5-Pro share?
19 benchmarks have published results for both models. GPT-4 Turbo has 36 scored results on Noometry and MiMo-V2.5-Pro has 27.