Model comparison
MiniMax-M2.7 vs Nova 2 Lite
Nova 2 Lite is the stronger model overall, scoring 39.0 to 37.4 on the Noometry Index. MiniMax-M2.7 costs 1.6× less per token, which makes it the better buy when Nova 2 Lite's lead doesn't matter for your workload.
Last verified . 18 shared benchmarks.
MiniMaxSummary
- They share 18 benchmarks with published results for both. MiniMax-M2.7 scores higher in 6 categories and Nova 2 Lite in 3 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where Nova 2 Lite leads 34.8 to 24.9.
- The biggest single-benchmark swing is Vectara Hallucination Rate: 12.9% for MiniMax-M2.7 and 5.1% for Nova 2 Lite.
- MiniMax-M2.7 is cheaper at $0.30 / $1.20 per million input/output tokens, against $0.30 / $2.50 for Nova 2 Lite.
- Nova 2 Lite accepts more context: 1M tokens versus 205K.
- MiniMax-M2.7 has downloadable open weights; the other is API-only.
Side by side
| MiniMax-M2.7 | Nova 2 Lite | |
|---|---|---|
| Provider | MiniMax | Amazon |
| Noometry Index | 37.4 | 39.0 |
| Released | 2026-03-18 | 2025-12-01 |
| Weights | Open | Proprietary |
| Context window | 205K | 1M |
| Max output | 131K | 64K |
| Input $ / M tokens | $0.30 | $0.30 |
| Output $ / M tokens | $1.20 | $2.50 |
| Results tracked | 30 | 19 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding MiniMax-M2.7 leads
MiniMax-M2.7: 41.4 (#115), Nova 2 Lite: 40.1 (#131)
| Benchmark | MiniMax-M2.7 | Nova 2 Lite |
|---|---|---|
| LMArena Coding | 1454 | 1386 |
| LMArena WebDev | 1397 | — |
| SciCode | 47% | — |
| WeirdML | 37% | — |
| ALE-Bench | 599.25 | — |
Agentic & Tool Use MiniMax-M2.7 leads
MiniMax-M2.7: 24.9 (#103), Nova 2 Lite: 23.9 (#114)
| Benchmark | MiniMax-M2.7 | Nova 2 Lite |
|---|---|---|
| Terminal-Bench | 45.1% | — |
| Berkeley Function Calling Leaderboard | — | 27.1% |
| ExploitBench | 13.3% | — |
| GBAEval | 0% | — |
Reasoning Nova 2 Lite leads
MiniMax-M2.7: 19.0 (#248), Nova 2 Lite: 26.4 (#120)
| Benchmark | MiniMax-M2.7 | Nova 2 Lite |
|---|---|---|
| LMArena Hard Prompts | 1422 | 1364 |
| NYT Connections (extended) | 24.7% | — |
| CritPt | 0.6% | — |
| Thematic Generalization | 39.3% | — |
| Epoch Capabilities Index | 145.84 | — |
Math Nova 2 Lite leads
MiniMax-M2.7: 24.9 (#254), Nova 2 Lite: 34.8 (#154)
| Benchmark | MiniMax-M2.7 | Nova 2 Lite |
|---|---|---|
| LMArena Math | 1420 | 1359 |
| ProofBench | 3% | — |
Knowledge Nova 2 Lite leads
MiniMax-M2.7: 37.1 (#151), Nova 2 Lite: 42.5 (#88)
| Benchmark | MiniMax-M2.7 | Nova 2 Lite |
|---|---|---|
| Vectara Hallucination Rate | 12.9% | 5.1% |
| LMArena Expert | 1443 | 1360 |
Multilingual MiniMax-M2.7 leads
MiniMax-M2.7: 50.5 (#120), Nova 2 Lite: 47.3 (#148)
| Benchmark | MiniMax-M2.7 | Nova 2 Lite |
|---|---|---|
| LMArena Non-English | 1382 | 1337 |
| LMArena Chinese | 1439 | 1364 |
| LMArena French | 1422 | 1381 |
| LMArena German | 1398 | 1343 |
| LMArena Japanese | 1261 | 1273 |
| LMArena Korean | 1313 | 1284 |
| LMArena Russian | 1383 | 1343 |
| LMArena Spanish | 1403 | 1372 |
Instruction Following MiniMax-M2.7 leads
MiniMax-M2.7: 74.2 (#101), Nova 2 Lite: 70.7 (#156)
| Benchmark | MiniMax-M2.7 | Nova 2 Lite |
|---|---|---|
| LMArena Instruction Following | 1404 | 1335 |
Long Context MiniMax-M2.7 leads
MiniMax-M2.7: 43.3 (#94), Nova 2 Lite: 40.6 (#145)
| Benchmark | MiniMax-M2.7 | Nova 2 Lite |
|---|---|---|
| LMArena Longer Query | 1419 | 1335 |
Writing & Preference MiniMax-M2.7 leads
MiniMax-M2.7: 59.1 (#107), Nova 2 Lite: 54.1 (#150)
| Benchmark | MiniMax-M2.7 | Nova 2 Lite |
|---|---|---|
| LMArena Text | 1404 | 1362 |
| LMArena Creative Writing | 1354 | 1291 |
| LMArena Multi-Turn | 1413 | 1338 |
Frequently asked questions
Is MiniMax-M2.7 better than Nova 2 Lite?
Nova 2 Lite is the stronger model overall, scoring 39.0 to 37.4 on the Noometry Index. MiniMax-M2.7 costs 1.6× less per token, which makes it the better buy when Nova 2 Lite's lead doesn't matter for your workload.
Which is cheaper, MiniMax-M2.7 or Nova 2 Lite?
MiniMax-M2.7 is cheaper. It lists at $0.30 per million input tokens and $1.20 per million output tokens; Nova 2 Lite lists at $0.30 and $2.50.
Is MiniMax-M2.7 or Nova 2 Lite better for coding?
MiniMax-M2.7 scores higher on coding benchmarks: 41.4 versus 40.1 in the Noometry coding category.
Which has the bigger context window?
Nova 2 Lite does, with 1M tokens against 205K.
How many benchmarks do MiniMax-M2.7 and Nova 2 Lite share?
18 benchmarks have published results for both models. MiniMax-M2.7 has 30 scored results on Noometry and Nova 2 Lite has 19.