Model comparison
Amazon Nova Micro vs MiniMax-M2.5
MiniMax-M2.5 is the stronger model overall, scoring 38.3 to 30.4 on the Noometry Index. Amazon Nova Micro costs 8.6× less per token, which makes it the better buy when MiniMax-M2.5's lead doesn't matter for your workload.
Last verified . 18 shared benchmarks.
MiniMaxSummary
- They share 18 benchmarks with published results for both. Amazon Nova Micro scores higher in 0 categories and MiniMax-M2.5 in 9 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in coding, where MiniMax-M2.5 leads 48.1 to 30.5.
- Amazon Nova Micro is cheaper at $0.035 / $0.14 per million input/output tokens, against $0.30 / $1.20 for MiniMax-M2.5.
- MiniMax-M2.5 accepts more context: 205K tokens versus 128K.
- MiniMax-M2.5 has downloadable open weights; the other is API-only.
Side by side
| Amazon Nova Micro | MiniMax-M2.5 | |
|---|---|---|
| Provider | Amazon | MiniMax |
| Noometry Index | 30.4 | 38.3 |
| Released | 2024-12-03 | 2026-02-12 |
| Weights | Proprietary | Open |
| Context window | 128K | 205K |
| Max output | 10K | 131K |
| Input $ / M tokens | $0.035 | $0.30 |
| Output $ / M tokens | $0.14 | $1.20 |
| Results tracked | 32 | 33 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding MiniMax-M2.5 leads
Amazon Nova Micro: 30.5 (#295), MiniMax-M2.5: 48.1 (#58)
| Benchmark | Amazon Nova Micro | MiniMax-M2.5 |
|---|---|---|
| LMArena Coding | 1218 | 1381 |
| SWE-bench Verified (bash only) | — | 75.8% |
| LMArena WebDev | — | 1387 |
| SWE-bench Multilingual | — | 68.3% |
| LiveBench Coding | 20.2% | — |
| ALE-Bench | — | 618.17 |
Agentic & Tool Use MiniMax-M2.5 leads
Amazon Nova Micro: 22.1 (#132), MiniMax-M2.5: 30.4 (#77)
| Benchmark | Amazon Nova Micro | MiniMax-M2.5 |
|---|---|---|
| Terminal-Bench | — | 42.7% |
| Berkeley Function Calling Leaderboard | 22.3% | — |
| Vending-Bench 2 | — | -23.16 |
Reasoning Too close to call
Amazon Nova Micro: 17.4 (#294), MiniMax-M2.5: 17.5 (#292)
| Benchmark | Amazon Nova Micro | MiniMax-M2.5 |
|---|---|---|
| LMArena Hard Prompts | 1191 | 1372 |
| ARC-AGI-2 | — | 4.9% |
| Kagi LLM Benchmark | — | 55.2% |
| NYT Connections (extended) | — | 16.8% |
| ARC-AGI-1 | — | 63.7% |
| LiveBench Reasoning | 25.1% | — |
| LiveBench Data Analysis | 34% | — |
| Epoch Capabilities Index | — | 146.68 |
| LiveBench | 29.6% | — |
Math Too close to call
Amazon Nova Micro: 26.9 (#254), MiniMax-M2.5: 26.9 (#253)
| Benchmark | Amazon Nova Micro | MiniMax-M2.5 |
|---|---|---|
| LMArena Math | 1206 | 1378 |
| ProofBench | — | 4% |
| Omni-MATH | 21.4% | — |
| LiveBench Math | 34.5% | — |
Knowledge MiniMax-M2.5 leads
Amazon Nova Micro: 29.6 (#237), MiniMax-M2.5: 39.2 (#135)
| Benchmark | Amazon Nova Micro | MiniMax-M2.5 |
|---|---|---|
| Vectara Hallucination Rate | 5.5% | 9.1% |
| LMArena Expert | 1184 | 1379 |
| MMLU-Pro | 51.1% | — |
| GPQA (HELM) | 38.3% | — |
| MMLU | 70.8% | — |
Multilingual MiniMax-M2.5 leads
Amazon Nova Micro: 36.5 (#239), MiniMax-M2.5: 47.1 (#152)
| Benchmark | Amazon Nova Micro | MiniMax-M2.5 |
|---|---|---|
| LMArena Non-English | 1186 | 1338 |
| LMArena Chinese | 1209 | 1393 |
| LMArena French | 1238 | 1362 |
| LMArena German | 1192 | 1362 |
| LMArena Japanese | 1154 | 1171 |
| LMArena Korean | 1150 | 1232 |
| LMArena Russian | 1185 | 1358 |
| LMArena Spanish | 1225 | 1354 |
Instruction Following MiniMax-M2.5 leads
Amazon Nova Micro: 56.3 (#272), MiniMax-M2.5: 71.5 (#148)
| Benchmark | Amazon Nova Micro | MiniMax-M2.5 |
|---|---|---|
| LMArena Instruction Following | 1174 | 1353 |
| LiveBench Instruction Following | 48% | — |
| IFEval | 76% | — |
Long Context Too close to call
Amazon Nova Micro: 36.5 (#229), MiniMax-M2.5: 37.5 (#216)
| Benchmark | Amazon Nova Micro | MiniMax-M2.5 |
|---|---|---|
| LMArena Longer Query | 1205 | 1366 |
| CL-bench | — | 11.4% |
| CL-bench Life | — | 6.3% |
Writing & Preference MiniMax-M2.5 leads
Amazon Nova Micro: 39.5 (#247), MiniMax-M2.5: 53.9 (#153)
| Benchmark | Amazon Nova Micro | MiniMax-M2.5 |
|---|---|---|
| LMArena Text | 1208 | 1359 |
| LMArena Creative Writing | 1172 | 1331 |
| LMArena Multi-Turn | 1178 | 1364 |
| EQ-Bench Creative Writing | — | 1361 |
| WildBench | 74.3% | — |
| LiveBench Language | 15.8% | — |
Frequently asked questions
Is Amazon Nova Micro better than MiniMax-M2.5?
MiniMax-M2.5 is the stronger model overall, scoring 38.3 to 30.4 on the Noometry Index. Amazon Nova Micro costs 8.6× less per token, which makes it the better buy when MiniMax-M2.5's lead doesn't matter for your workload.
Which is cheaper, Amazon Nova Micro or MiniMax-M2.5?
Amazon Nova Micro is cheaper. It lists at $0.035 per million input tokens and $0.14 per million output tokens; MiniMax-M2.5 lists at $0.30 and $1.20.
Is Amazon Nova Micro or MiniMax-M2.5 better for coding?
MiniMax-M2.5 scores higher on coding benchmarks: 48.1 versus 30.5 in the Noometry coding category.
Which has the bigger context window?
MiniMax-M2.5 does, with 205K tokens against 128K.
How many benchmarks do Amazon Nova Micro and MiniMax-M2.5 share?
18 benchmarks have published results for both models. Amazon Nova Micro has 32 scored results on Noometry and MiniMax-M2.5 has 33.