Model comparison
GLM-4.5-Air vs Grok-3 mini
Grok-3 mini is the stronger model overall, scoring 41.0 to 38.7 on the Noometry Index.
Last verified . 23 shared benchmarks.
Summary
- They share 23 benchmarks with published results for both. GLM-4.5-Air scores higher in 4 categories and Grok-3 mini in 4 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Grok-3 mini leads 46.3 to 34.6.
- The biggest single-benchmark swing is Kagi LLM Benchmark: 43% for GLM-4.5-Air and 61.3% for Grok-3 mini.
- GLM-4.5-Air has downloadable open weights; the other is API-only.
Side by side
| GLM-4.5-Air | Grok-3 mini | |
|---|---|---|
| Provider | Z.ai (Zhipu) | xAI |
| Noometry Index | 38.7 | 41.0 |
| Released | 2025-07-20 | 2025-04-09 |
| Weights | Open | Proprietary |
| Context window | 131K | — |
| Max output | 98K | — |
| Input $ / M tokens | $0.20 | — |
| Output $ / M tokens | $1.10 | — |
| Results tracked | 27 | 35 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Grok-3 mini leads
GLM-4.5-Air: 34.5 (#225), Grok-3 mini: 40.4 (#126)
| Benchmark | GLM-4.5-Air | Grok-3 mini |
|---|---|---|
| LMArena Coding | 1397 | 1380 |
| Aider Polyglot | — | 49.3% |
| GSO | 2.9% | — |
| WeirdML | — | 42.6% |
Reasoning GLM-4.5-Air leads
GLM-4.5-Air: 23.3 (#166), Grok-3 mini: 13.8 (#323)
| Benchmark | GLM-4.5-Air | Grok-3 mini |
|---|---|---|
| Kagi LLM Benchmark | 43% | 61.3% |
| LMArena Hard Prompts | 1379 | 1374 |
| ARC-AGI-2 | — | 0.4% |
| ARC-AGI-1 | — | 16.5% |
| Epoch Capabilities Index | — | 140.35 |
| ForecastBench | 59.2 | — |
Math Grok-3 mini leads
GLM-4.5-Air: 33.8 (#166), Grok-3 mini: 40.3 (#78)
| Benchmark | GLM-4.5-Air | Grok-3 mini |
|---|---|---|
| Omni-MATH | 39.1% | 31.8% |
| LMArena Math | 1396 | 1386 |
| OTIS Mock AIME 2024-2025 | — | 77.8% |
| MATH Level 5 | — | 90.9% |
| FrontierMath (Feb 2025 set) | — | 5.9% |
Knowledge Grok-3 mini leads
GLM-4.5-Air: 34.6 (#184), Grok-3 mini: 46.3 (#75)
| Benchmark | GLM-4.5-Air | Grok-3 mini |
|---|---|---|
| MMLU-Pro | 76.2% | 79.9% |
| GPQA (HELM) | 59.4% | 67.5% |
| LMArena Expert | 1370 | 1395 |
| GPQA Diamond | — | 76.3% |
| Humanity's Last Exam | 8.1% | — |
| Confabulations | — | 10.8% |
| Vectara Hallucination Rate | 9.3% | — |
Multilingual GLM-4.5-Air leads
GLM-4.5-Air: 49.4 (#130), Grok-3 mini: 48.3 (#141)
| Benchmark | GLM-4.5-Air | Grok-3 mini |
|---|---|---|
| LMArena Non-English | 1366 | 1352 |
| LMArena Chinese | 1427 | 1387 |
| LMArena French | 1399 | 1357 |
| LMArena German | 1376 | 1350 |
| LMArena Japanese | 1348 | 1343 |
| LMArena Korean | 1306 | 1336 |
| LMArena Russian | 1373 | 1352 |
| LMArena Spanish | 1385 | 1379 |
Instruction Following Grok-3 mini leads
GLM-4.5-Air: 69.8 (#165), Grok-3 mini: 78.7 (#8)
| Benchmark | GLM-4.5-Air | Grok-3 mini |
|---|---|---|
| IFEval | 81.2% | 95.1% |
| LMArena Instruction Following | 1353 | 1356 |
Long Context Too close to call
GLM-4.5-Air: 41.6 (#130), Grok-3 mini: 41.0 (#142)
| Benchmark | GLM-4.5-Air | Grok-3 mini |
|---|---|---|
| LMArena Longer Query | 1367 | 1372 |
| Fiction.LiveBench | — | 66.7% |
Writing & Preference GLM-4.5-Air leads
GLM-4.5-Air: 56.1 (#134), Grok-3 mini: 52.7 (#163)
| Benchmark | GLM-4.5-Air | Grok-3 mini |
|---|---|---|
| LMArena Text | 1384 | 1370 |
| LMArena Creative Writing | 1343 | 1342 |
| WildBench | 78.9% | 65.1% |
| LMArena Multi-Turn | 1371 | 1355 |
| Short-Story Creative Writing | — | 73.5% |
Frequently asked questions
Is GLM-4.5-Air better than Grok-3 mini?
Grok-3 mini is the stronger model overall, scoring 41.0 to 38.7 on the Noometry Index.
Is GLM-4.5-Air or Grok-3 mini better for coding?
Grok-3 mini scores higher on coding benchmarks: 40.4 versus 34.5 in the Noometry coding category.
How many benchmarks do GLM-4.5-Air and Grok-3 mini share?
23 benchmarks have published results for both models. GLM-4.5-Air has 27 scored results on Noometry and Grok-3 mini has 35.