Model comparison
GPT-5.4 mini vs MiMo-V2.5
GPT-5.4 mini is the stronger model overall, scoring 45.0 to 43.4 on the Noometry Index. MiMo-V2.5 costs 9.6× less per token, which makes it the better buy when GPT-5.4 mini's lead doesn't matter for your workload.
Last verified . 23 shared benchmarks.
Summary
- They share 23 benchmarks with published results for both. GPT-5.4 mini scores higher in 6 categories and MiMo-V2.5 in 3 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where GPT-5.4 mini leads 51.5 to 40.8.
- The biggest single-benchmark swing is SciCode: 49.9% for GPT-5.4 mini and 43.1% for MiMo-V2.5.
- MiMo-V2.5 is cheaper at $0.14 / $0.28 per million input/output tokens, against $0.75 / $4.50 for GPT-5.4 mini.
- MiMo-V2.5 accepts more context: 1.05M tokens versus 400K.
- MiMo-V2.5 has downloadable open weights; the other is API-only.
Side by side
| GPT-5.4 mini | MiMo-V2.5 | |
|---|---|---|
| Provider | OpenAI | Xiaomi |
| Noometry Index | 45.0 | 43.4 |
| Released | 2026-03-17 | 2026-04-22 |
| Weights | Proprietary | Open |
| Context window | 400K | 1.05M |
| Max output | 128K | 131K |
| Input $ / M tokens | $0.75 | $0.14 |
| Output $ / M tokens | $4.50 | $0.28 |
| Results tracked | 46 | 23 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5.4 mini leads
GPT-5.4 mini: 45.2 (#72), MiMo-V2.5: 43.9 (#81)
| Benchmark | GPT-5.4 mini | MiMo-V2.5 |
|---|---|---|
| LMArena WebDev | 1397 | 1438 |
| SciCode | 49.9% | 43.1% |
| LMArena Coding | 1438 | 1469 |
| ALE-Bench | 1,189 | 513.95 |
| FrontierCode | 27% | — |
| WeirdML | 60.3% | — |
Agentic & Tool Use Not comparable
GPT-5.4 mini: 29.9 (#81), MiMo-V2.5: —
| Benchmark | GPT-5.4 mini | MiMo-V2.5 |
|---|---|---|
| DeepResearch Bench | 36.3% | — |
Reasoning GPT-5.4 mini leads
GPT-5.4 mini: 30.4 (#85), MiMo-V2.5: 28.6 (#101)
| Benchmark | GPT-5.4 mini | MiMo-V2.5 |
|---|---|---|
| CritPt | 10% | 3.7% |
| LMArena Hard Prompts | 1424 | 1450 |
| ARC-AGI-2 | 18.9% | — |
| Kagi LLM Benchmark | 37.9% | — |
| NYT Connections (extended) | 61.8% | — |
| ARC-AGI-1 | 63.7% | — |
| Chess Puzzles | 24% | — |
| Thematic Generalization | 61.7% | — |
| Mystery Game Puzzles | 11% | — |
| DTBench | 80% | — |
| LMCA | 40.8% | — |
| Epoch Capabilities Index | 148.84 | — |
| ForecastBench | 57 | — |
Math GPT-5.4 mini leads
GPT-5.4 mini: 45.5 (#75), MiMo-V2.5: 36.8 (#163)
| Benchmark | GPT-5.4 mini | MiMo-V2.5 |
|---|---|---|
| ProofBench | 21% | 16% |
| LMArena Math | 1419 | 1436 |
| FrontierMath (Tiers 1-3) | 51.2% | — |
| FrontierMath Tier 4 | 9.8% | — |
| OTIS Mock AIME 2024-2025 | 88.9% | — |
| FrontierMath (Feb 2025 set) | 28.3% | — |
| FrontierMath Tier 4 (v1) | 2.1% | — |
Knowledge GPT-5.4 mini leads
GPT-5.4 mini: 51.5 (#67), MiMo-V2.5: 40.8 (#115)
| Benchmark | GPT-5.4 mini | MiMo-V2.5 |
|---|---|---|
| LMArena Expert | 1435 | 1460 |
| GPQA Diamond | 86.9% | — |
| SimpleQA Verified | 29.4% | — |
| Vectara Hallucination Rate | 5.5% | — |
Multimodal Too close to call
GPT-5.4 mini: 39.7 (#56), MiMo-V2.5: 39.8 (#54)
| Benchmark | GPT-5.4 mini | MiMo-V2.5 |
|---|---|---|
| LMArena Vision | 1245 | 1247 |
Multilingual Too close to call
GPT-5.4 mini: 51.9 (#96), MiMo-V2.5: 51.9 (#99)
| Benchmark | GPT-5.4 mini | MiMo-V2.5 |
|---|---|---|
| LMArena Non-English | 1405 | 1404 |
| LMArena Chinese | 1446 | 1468 |
| LMArena French | 1440 | 1447 |
| LMArena German | 1409 | 1421 |
| LMArena Japanese | 1374 | 1306 |
| LMArena Korean | 1368 | 1363 |
| LMArena Russian | 1417 | 1395 |
| LMArena Spanish | 1405 | 1416 |
Instruction Following MiMo-V2.5 leads
GPT-5.4 mini: 74.1 (#102), MiMo-V2.5: 75.5 (#60)
| Benchmark | GPT-5.4 mini | MiMo-V2.5 |
|---|---|---|
| LMArena Instruction Following | 1405 | 1434 |
Long Context MiMo-V2.5 leads
GPT-5.4 mini: 43.0 (#112), MiMo-V2.5: 44.2 (#73)
| Benchmark | GPT-5.4 mini | MiMo-V2.5 |
|---|---|---|
| LMArena Longer Query | 1407 | 1445 |
Writing & Preference GPT-5.4 mini leads
GPT-5.4 mini: 64.0 (#58), MiMo-V2.5: 61.6 (#86)
| Benchmark | GPT-5.4 mini | MiMo-V2.5 |
|---|---|---|
| LMArena Text | 1412 | 1428 |
| LMArena Creative Writing | 1370 | 1393 |
| LMArena Multi-Turn | 1429 | 1445 |
| EQ-Bench Creative Writing | 1665 | — |
Frequently asked questions
Is GPT-5.4 mini better than MiMo-V2.5?
GPT-5.4 mini is the stronger model overall, scoring 45.0 to 43.4 on the Noometry Index. MiMo-V2.5 costs 9.6× less per token, which makes it the better buy when GPT-5.4 mini's lead doesn't matter for your workload.
Which is cheaper, GPT-5.4 mini or MiMo-V2.5?
MiMo-V2.5 is cheaper. It lists at $0.14 per million input tokens and $0.28 per million output tokens; GPT-5.4 mini lists at $0.75 and $4.50.
Is GPT-5.4 mini or MiMo-V2.5 better for coding?
GPT-5.4 mini scores higher on coding benchmarks: 45.2 versus 43.9 in the Noometry coding category.
Which has the bigger context window?
MiMo-V2.5 does, with 1.05M tokens against 400K.
How many benchmarks do GPT-5.4 mini and MiMo-V2.5 share?
23 benchmarks have published results for both models. GPT-5.4 mini has 46 scored results on Noometry and MiMo-V2.5 has 23.