Model comparison
MiniMax-M2.5 vs Olmo 3 32b Think
MiniMax-M2.5 and Olmo 3 32b Think score almost the same on the Noometry Index (38.3 vs 38.7), so choose on price, context window or the category you care about most.
Last verified . 14 shared benchmarks.
MiniMax
Allen Institute for AI (Ai2)Summary
- They share 14 benchmarks with published results for both. MiniMax-M2.5 scores higher in 5 categories and Olmo 3 32b Think in 3 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Olmo 3 32b Think leads 36.5 to 26.9.
Side by side
| MiniMax-M2.5 | Olmo 3 32b Think | |
|---|---|---|
| Provider | MiniMax | Allen Institute for AI (Ai2) |
| Noometry Index | 38.3 | 38.7 |
| Released | 2026-02-12 | — |
| Weights | Open | Open |
| Context window | 205K | — |
| Max output | 131K | — |
| Input $ / M tokens | $0.30 | — |
| Output $ / M tokens | $1.20 | — |
| Results tracked | 33 | 14 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding MiniMax-M2.5 leads
MiniMax-M2.5: 48.1 (#58), Olmo 3 32b Think: 38.6 (#172)
| Benchmark | MiniMax-M2.5 | Olmo 3 32b Think |
|---|---|---|
| LMArena Coding | 1381 | 1319 |
| SWE-bench Verified (bash only) | 75.8% | — |
| LMArena WebDev | 1387 | — |
| SWE-bench Multilingual | 68.3% | — |
| ALE-Bench | 618.17 | — |
Agentic & Tool Use Not comparable
MiniMax-M2.5: 30.4 (#77), Olmo 3 32b Think: —
| Benchmark | MiniMax-M2.5 | Olmo 3 32b Think |
|---|---|---|
| Terminal-Bench | 42.7% | — |
| Vending-Bench 2 | -23.16 | — |
Reasoning Olmo 3 32b Think leads
MiniMax-M2.5: 17.5 (#292), Olmo 3 32b Think: 25.9 (#140)
| Benchmark | MiniMax-M2.5 | Olmo 3 32b Think |
|---|---|---|
| LMArena Hard Prompts | 1372 | 1302 |
| ARC-AGI-2 | 4.9% | — |
| Kagi LLM Benchmark | 55.2% | — |
| NYT Connections (extended) | 16.8% | — |
| ARC-AGI-1 | 63.7% | — |
| Epoch Capabilities Index | 146.68 | — |
Math Olmo 3 32b Think leads
MiniMax-M2.5: 26.9 (#253), Olmo 3 32b Think: 36.5 (#165)
| Benchmark | MiniMax-M2.5 | Olmo 3 32b Think |
|---|---|---|
| LMArena Math | 1378 | 1316 |
| ProofBench | 4% | — |
Knowledge MiniMax-M2.5 leads
MiniMax-M2.5: 39.2 (#135), Olmo 3 32b Think: 35.0 (#190)
| Benchmark | MiniMax-M2.5 | Olmo 3 32b Think |
|---|---|---|
| LMArena Expert | 1379 | 1273 |
| Vectara Hallucination Rate | 9.1% | — |
Multilingual MiniMax-M2.5 leads
MiniMax-M2.5: 47.1 (#152), Olmo 3 32b Think: 41.2 (#210)
| Benchmark | MiniMax-M2.5 | Olmo 3 32b Think |
|---|---|---|
| LMArena Non-English | 1338 | 1255 |
| LMArena Chinese | 1393 | 1300 |
| LMArena French | 1362 | 1291 |
| LMArena German | 1362 | 1290 |
| LMArena Russian | 1358 | 1254 |
| LMArena Japanese | 1171 | — |
| LMArena Korean | 1232 | — |
| LMArena Spanish | 1354 | — |
Instruction Following MiniMax-M2.5 leads
MiniMax-M2.5: 71.5 (#148), Olmo 3 32b Think: 67.2 (#198)
| Benchmark | MiniMax-M2.5 | Olmo 3 32b Think |
|---|---|---|
| LMArena Instruction Following | 1353 | 1275 |
Long Context Olmo 3 32b Think leads
MiniMax-M2.5: 37.5 (#216), Olmo 3 32b Think: 39.4 (#182)
| Benchmark | MiniMax-M2.5 | Olmo 3 32b Think |
|---|---|---|
| LMArena Longer Query | 1366 | 1296 |
| CL-bench | 11.4% | — |
| CL-bench Life | 6.3% | — |
Writing & Preference MiniMax-M2.5 leads
MiniMax-M2.5: 53.9 (#153), Olmo 3 32b Think: 49.1 (#193)
| Benchmark | MiniMax-M2.5 | Olmo 3 32b Think |
|---|---|---|
| LMArena Text | 1359 | 1300 |
| LMArena Creative Writing | 1331 | 1256 |
| LMArena Multi-Turn | 1364 | 1290 |
| EQ-Bench Creative Writing | 1361 | — |
Frequently asked questions
Is MiniMax-M2.5 better than Olmo 3 32b Think?
MiniMax-M2.5 and Olmo 3 32b Think score almost the same on the Noometry Index (38.3 vs 38.7), so choose on price, context window or the category you care about most.
Is MiniMax-M2.5 or Olmo 3 32b Think better for coding?
MiniMax-M2.5 scores higher on coding benchmarks: 48.1 versus 38.6 in the Noometry coding category.
How many benchmarks do MiniMax-M2.5 and Olmo 3 32b Think share?
14 benchmarks have published results for both models. MiniMax-M2.5 has 33 scored results on Noometry and Olmo 3 32b Think has 14.