Model comparison
gpt-oss-120b vs MiniMax-M2.5
MiniMax-M2.5 is the stronger model overall, scoring 38.3 to 36.3 on the Noometry Index. gpt-oss-120b costs 7.5× less per token, which makes it the better buy when MiniMax-M2.5's lead doesn't matter for your workload.
Last verified . 25 shared benchmarks.
MiniMaxSummary
- They share 25 benchmarks with published results for both. gpt-oss-120b scores higher in 4 categories and MiniMax-M2.5 in 5 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where gpt-oss-120b leads 52.5 to 26.9.
- The biggest single-benchmark swing is SWE-bench Verified (bash only): 26% for gpt-oss-120b and 75.8% for MiniMax-M2.5.
- gpt-oss-120b is cheaper at $0.037 / $0.17 per million input/output tokens, against $0.30 / $1.20 for MiniMax-M2.5.
- MiniMax-M2.5 accepts more context: 205K tokens versus 131K.
Side by side
| gpt-oss-120b | MiniMax-M2.5 | |
|---|---|---|
| Provider | OpenAI | MiniMax |
| Noometry Index | 36.3 | 38.3 |
| Released | 2025-08-05 | 2026-02-12 |
| Weights | Open | Open |
| Context window | 131K | 205K |
| Max output | 41K | 131K |
| Input $ / M tokens | $0.037 | $0.30 |
| Output $ / M tokens | $0.17 | $1.20 |
| Results tracked | 48 | 33 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding MiniMax-M2.5 leads
gpt-oss-120b: 33.5 (#256), MiniMax-M2.5: 48.1 (#58)
| Benchmark | gpt-oss-120b | MiniMax-M2.5 |
|---|---|---|
| SWE-bench Verified (bash only) | 26% | 75.8% |
| LMArena Coding | 1380 | 1381 |
| ALE-Bench | 575.62 | 618.17 |
| Aider Polyglot | 41.8% | — |
| LMArena WebDev | — | 1387 |
| SWE-bench Multilingual | — | 68.3% |
| SciCode | 36% | — |
| WeirdML | 48.2% | — |
| AlgoTune | 1.41 | — |
Agentic & Tool Use MiniMax-M2.5 leads
gpt-oss-120b: 12.2 (#153), MiniMax-M2.5: 30.4 (#77)
| Benchmark | gpt-oss-120b | MiniMax-M2.5 |
|---|---|---|
| Terminal-Bench | 18.7% | 42.7% |
| Vending-Bench 2 | -21.53 | -23.16 |
| APEX-Agents | 4.4% | — |
| METR Time Horizons | 56.6% | — |
Reasoning gpt-oss-120b leads
gpt-oss-120b: 20.0 (#245), MiniMax-M2.5: 17.5 (#292)
| Benchmark | gpt-oss-120b | MiniMax-M2.5 |
|---|---|---|
| Kagi LLM Benchmark | 58.6% | 55.2% |
| LMArena Hard Prompts | 1364 | 1372 |
| Epoch Capabilities Index | 139.93 | 146.68 |
| ARC-AGI-2 | — | 4.9% |
| SimpleBench | 22.1% | — |
| NYT Connections (extended) | — | 16.8% |
| ARC-AGI-1 | — | 63.7% |
| CritPt | 1.1% | — |
| Chess Puzzles | 20% | — |
| Mystery Game Puzzles | 2% | — |
| DTBench | 76.3% | — |
| LMCA | 22.1% | — |
| Surface Evolver Bench | 25% | — |
Math gpt-oss-120b leads
gpt-oss-120b: 52.5 (#50), MiniMax-M2.5: 26.9 (#253)
| Benchmark | gpt-oss-120b | MiniMax-M2.5 |
|---|---|---|
| LMArena Math | 1389 | 1378 |
| OTIS Mock AIME 2024-2025 | 88.9% | — |
| ProofBench | — | 4% |
| Omni-MATH | 68.8% | — |
Knowledge gpt-oss-120b leads
gpt-oss-120b: 42.4 (#96), MiniMax-M2.5: 39.2 (#135)
| Benchmark | gpt-oss-120b | MiniMax-M2.5 |
|---|---|---|
| Vectara Hallucination Rate | 14.2% | 9.1% |
| LMArena Expert | 1356 | 1379 |
| GPQA Diamond | 75.8% | — |
| MMLU-Pro | 79.5% | — |
| Confabulations | 15.7% | — |
| GPQA (HELM) | 68.4% | — |
Multilingual Too close to call
gpt-oss-120b: 48.0 (#147), MiniMax-M2.5: 47.1 (#152)
| Benchmark | gpt-oss-120b | MiniMax-M2.5 |
|---|---|---|
| LMArena Non-English | 1351 | 1338 |
| LMArena Chinese | 1385 | 1393 |
| LMArena French | 1369 | 1362 |
| LMArena German | 1353 | 1362 |
| LMArena Japanese | 1331 | 1171 |
| LMArena Korean | 1282 | 1232 |
| LMArena Russian | 1343 | 1358 |
| LMArena Spanish | 1389 | 1354 |
Instruction Following MiniMax-M2.5 leads
gpt-oss-120b: 69.3 (#173), MiniMax-M2.5: 71.5 (#148)
| Benchmark | gpt-oss-120b | MiniMax-M2.5 |
|---|---|---|
| LMArena Instruction Following | 1318 | 1353 |
| IFEval | 83.6% | — |
Long Context MiniMax-M2.5 leads
gpt-oss-120b: 31.4 (#278), MiniMax-M2.5: 37.5 (#216)
| Benchmark | gpt-oss-120b | MiniMax-M2.5 |
|---|---|---|
| LMArena Longer Query | 1319 | 1366 |
| Fiction.LiveBench | 44.4% | — |
| CL-bench | — | 11.4% |
| CL-bench Life | — | 6.3% |
Writing & Preference MiniMax-M2.5 leads
gpt-oss-120b: 46.5 (#217), MiniMax-M2.5: 53.9 (#153)
| Benchmark | gpt-oss-120b | MiniMax-M2.5 |
|---|---|---|
| LMArena Text | 1365 | 1359 |
| LMArena Creative Writing | 1275 | 1331 |
| EQ-Bench Creative Writing | 961 | 1361 |
| LMArena Multi-Turn | 1340 | 1364 |
| Short-Story Creative Writing | 77.1% | — |
| WildBench | 84.5% | — |
Frequently asked questions
Is gpt-oss-120b better than MiniMax-M2.5?
MiniMax-M2.5 is the stronger model overall, scoring 38.3 to 36.3 on the Noometry Index. gpt-oss-120b costs 7.5× less per token, which makes it the better buy when MiniMax-M2.5's lead doesn't matter for your workload.
Which is cheaper, gpt-oss-120b or MiniMax-M2.5?
gpt-oss-120b is cheaper. It lists at $0.037 per million input tokens and $0.17 per million output tokens; MiniMax-M2.5 lists at $0.30 and $1.20.
Is gpt-oss-120b or MiniMax-M2.5 better for coding?
MiniMax-M2.5 scores higher on coding benchmarks: 48.1 versus 33.5 in the Noometry coding category.
Which has the bigger context window?
MiniMax-M2.5 does, with 205K tokens against 131K.
How many benchmarks do gpt-oss-120b and MiniMax-M2.5 share?
25 benchmarks have published results for both models. gpt-oss-120b has 48 scored results on Noometry and MiniMax-M2.5 has 33.