Model comparison
GLM-5 vs MiniMax-M3
GLM-5 is the stronger model overall, scoring 46.1 to 43.8 on the Noometry Index. MiniMax-M3 costs 3.0× less per token, which makes it the better buy when GLM-5's lead doesn't matter for your workload.
Last verified . 27 shared benchmarks.
MiniMaxSummary
- They share 27 benchmarks with published results for both. GLM-5 scores higher in 6 categories and MiniMax-M3 in 3 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in agentic & tool use, where GLM-5 leads 31.1 to 22.6.
- The biggest single-benchmark swing is NYT Connections (extended): 74.8% for GLM-5 and 65.1% for MiniMax-M3.
- MiniMax-M3 is cheaper at $0.30 / $1.20 per million input/output tokens, against $1 / $3.20 for GLM-5.
- MiniMax-M3 accepts more context: 1M tokens versus 205K.
Side by side
| GLM-5 | MiniMax-M3 | |
|---|---|---|
| Provider | Z.ai (Zhipu) | MiniMax |
| Noometry Index | 46.1 | 43.8 |
| Released | 2026-02-11 | 2026-06-01 |
| Weights | Open | Open |
| Context window | 205K | 1M |
| Max output | 131K | 512K |
| Input $ / M tokens | $1 | $0.30 |
| Output $ / M tokens | $3.20 | $1.20 |
| Results tracked | 45 | 41 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GLM-5 leads
GLM-5: 49.0 (#52), MiniMax-M3: 41.8 (#118)
| Benchmark | GLM-5 | MiniMax-M3 |
|---|---|---|
| LMArena WebDev | 1434 | 1482 |
| LMArena Coding | 1461 | 1469 |
| ALE-Bench | 765.62 | 640.02 |
| SWE-bench Verified | 72.1% | — |
| FrontierCode | — | 14.7% |
| SWE-bench Verified (bash only) | 72.8% | — |
| SWE-bench Multilingual | 69.7% | — |
| SciCode | — | 47.1% |
| WeirdML | 48.2% | — |
Agentic & Tool Use GLM-5 leads
GLM-5: 31.1 (#71), MiniMax-M3: 22.6 (#130)
| Benchmark | GLM-5 | MiniMax-M3 |
|---|---|---|
| Vending-Bench 2 | 4,432 | 2,158 |
| Terminal-Bench | 52.4% | — |
| APEX-Agents | — | 37.7% |
| OSWorld 2.0 | — | 4.6% |
| τ²-bench Airline | 82.5% | — |
| τ²-bench Banking | 9.8% | — |
| τ²-bench Retail | 73.7% | — |
| τ²-bench Telecom | 86.8% | — |
| GBAEval | — | 0.9% |
Reasoning MiniMax-M3 leads
GLM-5: 27.6 (#116), MiniMax-M3: 30.1 (#87)
| Benchmark | GLM-5 | MiniMax-M3 |
|---|---|---|
| SimpleBench | 53.2% | 45.8% |
| NYT Connections (extended) | 74.8% | 65.1% |
| Chess Puzzles | 10% | 14% |
| LMArena Hard Prompts | 1452 | 1447 |
| Epoch Capabilities Index | 145.83 | 146.95 |
| ForecastBench | 61 | 61.4 |
| ARC-AGI-2 | 4.9% | — |
| Kagi LLM Benchmark | 75% | — |
| ARC-AGI-1 | 44.7% | — |
| CritPt | — | 3.7% |
| Mystery Game Puzzles | — | 8% |
| DTBench | — | 78.9% |
| LMCA | — | 33.7% |
| Surface Evolver Bench | — | 55% |
Math GLM-5 leads
GLM-5: 46.4 (#71), MiniMax-M3: 40.0 (#95)
| Benchmark | GLM-5 | MiniMax-M3 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 80% | 71.1% |
| LMArena Math | 1440 | 1429 |
| MathArena Final-Answer Competitions | 65.7% | — |
| ProofBench | — | 18% |
| FrontierMath (Feb 2025 set) | 16.4% | — |
| FrontierMath Tier 4 (v1) | 2.1% | — |
Knowledge MiniMax-M3 leads
GLM-5: 52.3 (#64), MiniMax-M3: 58.4 (#35)
| Benchmark | GLM-5 | MiniMax-M3 |
|---|---|---|
| GPQA Diamond | 87.8% | 90.9% |
| LMArena Expert | 1454 | 1461 |
| Vectara Hallucination Rate | 10.1% | — |
Multimodal Not comparable
GLM-5: —, MiniMax-M3: 40.2 (#51)
| Benchmark | GLM-5 | MiniMax-M3 |
|---|---|---|
| LMArena Vision | — | 1253 |
| LMArena Document | — | 1435 |
Multilingual Too close to call
GLM-5: 53.7 (#58), MiniMax-M3: 53.0 (#75)
| Benchmark | GLM-5 | MiniMax-M3 |
|---|---|---|
| LMArena Non-English | 1430 | 1420 |
| LMArena Chinese | 1511 | 1463 |
| LMArena French | 1455 | 1447 |
| LMArena German | 1445 | 1426 |
| LMArena Japanese | 1416 | 1381 |
| LMArena Korean | 1423 | 1372 |
| LMArena Russian | 1436 | 1428 |
| LMArena Spanish | 1454 | 1432 |
Instruction Following Too close to call
GLM-5: 75.2 (#67), MiniMax-M3: 75.5 (#62)
| Benchmark | GLM-5 | MiniMax-M3 |
|---|---|---|
| LMArena Instruction Following | 1428 | 1433 |
Long Context Too close to call
GLM-5: 44.7 (#60), MiniMax-M3: 44.2 (#72)
| Benchmark | GLM-5 | MiniMax-M3 |
|---|---|---|
| LMArena Longer Query | 1446 | 1445 |
| CL-bench | 18.7% | — |
Writing & Preference GLM-5 leads
GLM-5: 66.0 (#38), MiniMax-M3: 62.1 (#83)
| Benchmark | GLM-5 | MiniMax-M3 |
|---|---|---|
| LMArena Text | 1446 | 1433 |
| LMArena Creative Writing | 1439 | 1404 |
| LMArena Multi-Turn | 1456 | 1442 |
| EQ-Bench Creative Writing | 1601 | — |
| EQ-Bench 4 | — | 1150 |
Frequently asked questions
Is GLM-5 better than MiniMax-M3?
GLM-5 is the stronger model overall, scoring 46.1 to 43.8 on the Noometry Index. MiniMax-M3 costs 3.0× less per token, which makes it the better buy when GLM-5's lead doesn't matter for your workload.
Which is cheaper, GLM-5 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 lists at $1 and $3.20.
Is GLM-5 or MiniMax-M3 better for coding?
GLM-5 scores higher on coding benchmarks: 49.0 versus 41.8 in the Noometry coding category.
Which has the bigger context window?
MiniMax-M3 does, with 1M tokens against 205K.
How many benchmarks do GLM-5 and MiniMax-M3 share?
27 benchmarks have published results for both models. GLM-5 has 45 scored results on Noometry and MiniMax-M3 has 41.