Model comparison
gpt-oss-20b vs MiMo-V2.5
MiMo-V2.5 is the stronger model overall, scoring 43.0 to 32.1 on the Noometry Index. gpt-oss-20b costs 4.9× less per token, which makes it the better buy when MiMo-V2.5's lead doesn't matter for your workload.
Last verified . 19 shared benchmarks.
Summary
- They share 19 benchmarks with published results for both. gpt-oss-20b scores higher in 1 category and MiMo-V2.5 in 7 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where MiMo-V2.5 leads 61.8 to 35.8.
- The biggest single-benchmark swing is SciCode: 34.4% for gpt-oss-20b and 43.1% for MiMo-V2.5.
- gpt-oss-20b is cheaper at $0.018 / $0.09 per million input/output tokens, against $0.14 / $0.28 for MiMo-V2.5.
- MiMo-V2.5 accepts more context: 1.05M tokens versus 131K.
Side by side
| gpt-oss-20b | MiMo-V2.5 | |
|---|---|---|
| Provider | OpenAI | Xiaomi |
| Noometry Index | 32.1 | 43.0 |
| Released | 2025-08-05 | 2026-04-22 |
| Weights | Open | Open |
| Context window | 131K | 1.05M |
| Max output | 16K | 131K |
| Input $ / M tokens | $0.018 | $0.14 |
| Output $ / M tokens | $0.09 | $0.28 |
| Results tracked | 34 | 23 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding MiMo-V2.5 leads
gpt-oss-20b: 37.1 (#185), MiMo-V2.5: 43.6 (#75)
| Benchmark | gpt-oss-20b | MiMo-V2.5 |
|---|---|---|
| SciCode | 34.4% | 43.1% |
| LMArena Coding | 1307 | 1469 |
| ALE-Bench | 566.05 | 513.95 |
| LMArena WebDev | — | 1438 |
| WeirdML | 40.9% | — |
Agentic & Tool Use Not comparable
gpt-oss-20b: 9.2 (#146), MiMo-V2.5: —
| Benchmark | gpt-oss-20b | MiMo-V2.5 |
|---|---|---|
| Terminal-Bench | 3.4% | — |
Reasoning MiMo-V2.5 leads
gpt-oss-20b: 18.8 (#251), MiMo-V2.5: 27.8 (#98)
| Benchmark | gpt-oss-20b | MiMo-V2.5 |
|---|---|---|
| CritPt | 1.4% | 3.7% |
| LMArena Hard Prompts | 1274 | 1450 |
| Kagi LLM Benchmark | 53.2% | — |
| Chess Puzzles | 4% | — |
| DTBench | 68% | — |
| LMCA | 14.5% | — |
| Epoch Capabilities Index | 137.81 | — |
Math gpt-oss-20b leads
gpt-oss-20b: 37.6 (#93), MiMo-V2.5: 35.4 (#145)
| Benchmark | gpt-oss-20b | MiMo-V2.5 |
|---|---|---|
| LMArena Math | 1317 | 1436 |
| OTIS Mock AIME 2024-2025 | 65.3% | — |
| ProofBench | — | 16% |
| Omni-MATH | 56.5% | — |
Knowledge MiMo-V2.5 leads
gpt-oss-20b: 34.6 (#186), MiMo-V2.5: 40.2 (#109)
| Benchmark | gpt-oss-20b | MiMo-V2.5 |
|---|---|---|
| LMArena Expert | 1258 | 1459 |
| GPQA Diamond | 60.8% | — |
| MMLU-Pro | 74% | — |
| GPQA (HELM) | 59.4% | — |
Multimodal Not comparable
gpt-oss-20b: —, MiMo-V2.5: 39.0 (#53)
| Benchmark | gpt-oss-20b | MiMo-V2.5 |
|---|---|---|
| LMArena Vision | — | 1245 |
Multilingual MiMo-V2.5 leads
gpt-oss-20b: 42.4 (#192), MiMo-V2.5: 52.0 (#96)
| Benchmark | gpt-oss-20b | MiMo-V2.5 |
|---|---|---|
| LMArena Non-English | 1269 | 1404 |
| LMArena Chinese | 1311 | 1467 |
| LMArena German | 1249 | 1421 |
| LMArena Japanese | 1245 | 1304 |
| LMArena Korean | 1239 | 1363 |
| LMArena Russian | 1279 | 1395 |
| LMArena Spanish | 1267 | 1412 |
| LMArena French | — | 1447 |
Instruction Following MiMo-V2.5 leads
gpt-oss-20b: 62.0 (#235), MiMo-V2.5: 75.7 (#55)
| Benchmark | gpt-oss-20b | MiMo-V2.5 |
|---|---|---|
| LMArena Instruction Following | 1236 | 1433 |
| IFEval | 73.2% | — |
Long Context MiMo-V2.5 leads
gpt-oss-20b: 37.9 (#203), MiMo-V2.5: 44.1 (#69)
| Benchmark | gpt-oss-20b | MiMo-V2.5 |
|---|---|---|
| LMArena Longer Query | 1251 | 1444 |
Writing & Preference MiMo-V2.5 leads
gpt-oss-20b: 35.8 (#260), MiMo-V2.5: 61.8 (#82)
| Benchmark | gpt-oss-20b | MiMo-V2.5 |
|---|---|---|
| LMArena Text | 1288 | 1427 |
| LMArena Creative Writing | 1201 | 1392 |
| LMArena Multi-Turn | 1268 | 1445 |
| EQ-Bench Creative Writing | 666 | — |
| WildBench | 73.7% | — |
Frequently asked questions
Is gpt-oss-20b better than MiMo-V2.5?
MiMo-V2.5 is the stronger model overall, scoring 43.0 to 32.1 on the Noometry Index. gpt-oss-20b costs 4.9× less per token, which makes it the better buy when MiMo-V2.5's lead doesn't matter for your workload.
Which is cheaper, gpt-oss-20b or MiMo-V2.5?
gpt-oss-20b is cheaper. It lists at $0.018 per million input tokens and $0.09 per million output tokens; MiMo-V2.5 lists at $0.14 and $0.28.
Is gpt-oss-20b or MiMo-V2.5 better for coding?
MiMo-V2.5 scores higher on coding benchmarks: 43.6 versus 37.1 in the Noometry coding category.
Which has the bigger context window?
MiMo-V2.5 does, with 1.05M tokens against 131K.
How many benchmarks do gpt-oss-20b and MiMo-V2.5 share?
19 benchmarks have published results for both models. gpt-oss-20b has 34 scored results on Noometry and MiMo-V2.5 has 23.