Model comparison
GLM-5.1 vs MiniMax-M3
GLM-5.1 is the stronger model overall, scoring 47.6 to 43.5 on the Noometry Index. MiniMax-M3 costs 4.1× less per token, which makes it the better buy when GLM-5.1's lead doesn't matter for your workload.
Last verified . 31 shared benchmarks.
MiniMaxSummary
- They share 31 benchmarks with published results for both. GLM-5.1 scores higher in 8 categories and MiniMax-M3 in 1 category; 7 gaps are clear of the uncertainty.
- The widest gap is in math, where GLM-5.1 leads 49.0 to 39.1.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 93.3% for GLM-5.1 and 71.1% for MiniMax-M3.
- MiniMax-M3 is cheaper at $0.30 / $1.20 per million input/output tokens, against $1.40 / $4.40 for GLM-5.1.
- MiniMax-M3 accepts more context: 1M tokens versus 200K.
Side by side
| GLM-5.1 | MiniMax-M3 | |
|---|---|---|
| Provider | Z.ai (Zhipu) | MiniMax |
| Noometry Index | 47.6 | 43.5 |
| Released | 2026-04-07 | 2026-06-01 |
| Weights | Open | Open |
| Context window | 200K | 1M |
| Max output | 131K | 512K |
| Input $ / M tokens | $1.40 | $0.30 |
| Output $ / M tokens | $4.40 | $1.20 |
| Results tracked | 41 | 41 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GLM-5.1 leads
GLM-5.1: 48.5 (#52), MiniMax-M3: 41.5 (#111)
| Benchmark | GLM-5.1 | MiniMax-M3 |
|---|---|---|
| LMArena WebDev | 1509 | 1482 |
| SciCode | 43.8% | 47.1% |
| LMArena Coding | 1486 | 1470 |
| ALE-Bench | 887.1 | 640.02 |
| SWE-bench Verified | 74.2% | — |
| FrontierCode | — | 14.7% |
| WeirdML | 57.1% | — |
Agentic & Tool Use GLM-5.1 leads
GLM-5.1: 24.8 (#106), MiniMax-M3: 22.5 (#121)
| Benchmark | GLM-5.1 | MiniMax-M3 |
|---|---|---|
| APEX-Agents | 40.9% | 37.7% |
| GBAEval | 0% | 0.9% |
| Vending-Bench 2 | 5,634 | 2,158 |
| OSWorld 2.0 | — | 4.6% |
| ExploitBench | 18.1% | — |
Reasoning GLM-5.1 leads
GLM-5.1: 38.2 (#57), MiniMax-M3: 29.3 (#83)
| Benchmark | GLM-5.1 | MiniMax-M3 |
|---|---|---|
| SimpleBench | 55.1% | 45.8% |
| NYT Connections (extended) | 77.7% | 65.1% |
| CritPt | 4.6% | 3.7% |
| Chess Puzzles | 19% | 14% |
| LMArena Hard Prompts | 1472 | 1447 |
| Epoch Capabilities Index | 149.86 | 147 |
| Thematic Generalization | 69.8% | — |
| Mystery Game Puzzles | — | 8% |
| DTBench | — | 78.9% |
| LMCA | — | 33.7% |
| Surface Evolver Bench | — | 55% |
| ForecastBench | — | 61.4 |
Math GLM-5.1 leads
GLM-5.1: 49.0 (#48), MiniMax-M3: 39.1 (#83)
| Benchmark | GLM-5.1 | MiniMax-M3 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 93.3% | 71.1% |
| ProofBench | 22.2% | 18% |
| LMArena Math | 1472 | 1432 |
| FrontierMath (Tiers 1-3) | 36.8% | — |
| MathArena Final-Answer Competitions | 67.1% | — |
| FrontierMath (Feb 2025 set) | 33.4% | — |
| FrontierMath Tier 4 (v1) | 12.5% | — |
Knowledge MiniMax-M3 leads
GLM-5.1: 55.0 (#45), MiniMax-M3: 58.4 (#32)
| Benchmark | GLM-5.1 | MiniMax-M3 |
|---|---|---|
| GPQA Diamond | 89.9% | 90.9% |
| LMArena Expert | 1477 | 1462 |
| SimpleQA Verified | 34% | — |
Multimodal Not comparable
GLM-5.1: —, MiniMax-M3: 39.5 (#45)
| Benchmark | GLM-5.1 | MiniMax-M3 |
|---|---|---|
| LMArena Vision | — | 1254 |
| LMArena Document | — | 1435 |
Multilingual GLM-5.1 leads
GLM-5.1: 55.2 (#34), MiniMax-M3: 53.2 (#70)
| Benchmark | GLM-5.1 | MiniMax-M3 |
|---|---|---|
| LMArena Non-English | 1448 | 1420 |
| LMArena Chinese | 1516 | 1465 |
| LMArena French | 1478 | 1445 |
| LMArena German | 1466 | 1424 |
| LMArena Japanese | 1436 | 1381 |
| LMArena Korean | 1423 | 1372 |
| LMArena Russian | 1453 | 1430 |
| LMArena Spanish | 1467 | 1431 |
Instruction Following Too close to call
GLM-5.1: 76.5 (#41), MiniMax-M3: 75.7 (#56)
| Benchmark | GLM-5.1 | MiniMax-M3 |
|---|---|---|
| LMArena Instruction Following | 1451 | 1433 |
Long Context Too close to call
GLM-5.1: 44.9 (#51), MiniMax-M3: 44.2 (#68)
| Benchmark | GLM-5.1 | MiniMax-M3 |
|---|---|---|
| LMArena Longer Query | 1467 | 1445 |
Writing & Preference GLM-5.1 leads
GLM-5.1: 67.2 (#29), MiniMax-M3: 62.3 (#79)
| Benchmark | GLM-5.1 | MiniMax-M3 |
|---|---|---|
| LMArena Text | 1462 | 1433 |
| LMArena Creative Writing | 1454 | 1405 |
| LMArena Multi-Turn | 1473 | 1441 |
| EQ-Bench Creative Writing | 1592 | — |
| EQ-Bench 4 | — | 1150 |
Frequently asked questions
Is GLM-5.1 better than MiniMax-M3?
GLM-5.1 is the stronger model overall, scoring 47.6 to 43.5 on the Noometry Index. MiniMax-M3 costs 4.1× less per token, which makes it the better buy when GLM-5.1's lead doesn't matter for your workload.
Which is cheaper, GLM-5.1 or MiniMax-M3?
MiniMax-M3 is cheaper. It lists at $0.30 per million input tokens and $1.20 per million output tokens; GLM-5.1 lists at $1.40 and $4.40.
Is GLM-5.1 or MiniMax-M3 better for coding?
GLM-5.1 scores higher on coding benchmarks: 48.5 versus 41.5 in the Noometry coding category.
Which has the bigger context window?
MiniMax-M3 does, with 1M tokens against 200K.
How many benchmarks do GLM-5.1 and MiniMax-M3 share?
31 benchmarks have published results for both models. GLM-5.1 has 41 scored results on Noometry and MiniMax-M3 has 41.