Model comparison
GLM-4.5-Air vs Grok 4.5
Grok 4.5 is the stronger model overall, scoring 55.0 to 38.7 on the Noometry Index. GLM-4.5-Air costs 7.1× less per token, which makes it the better buy when Grok 4.5's lead doesn't matter for your workload.
Last verified . 18 shared benchmarks.
Summary
- They share 18 benchmarks with published results for both. GLM-4.5-Air scores higher in 0 categories and Grok 4.5 in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Grok 4.5 leads 56.0 to 23.3.
- The biggest single-benchmark swing is Kagi LLM Benchmark: 43% for GLM-4.5-Air and 83.5% for Grok 4.5.
- GLM-4.5-Air is cheaper at $0.20 / $1.10 per million input/output tokens, against $2 / $6 for Grok 4.5.
- Grok 4.5 accepts more context: 500K tokens versus 131K.
- GLM-4.5-Air has downloadable open weights; the other is API-only.
Side by side
| GLM-4.5-Air | Grok 4.5 | |
|---|---|---|
| Provider | Z.ai (Zhipu) | xAI |
| Noometry Index | 38.7 | 55.0 |
| Released | 2025-07-20 | 2026-07-08 |
| Weights | Open | Proprietary |
| Context window | 131K | 500K |
| Max output | 98K | 500K |
| Input $ / M tokens | $0.20 | $2 |
| Output $ / M tokens | $1.10 | $6 |
| Results tracked | 27 | 52 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Grok 4.5 leads
GLM-4.5-Air: 34.5 (#225), Grok 4.5: 51.7 (#34)
| Benchmark | GLM-4.5-Air | Grok 4.5 |
|---|---|---|
| LMArena Coding | 1397 | 1473 |
| DeepSWE | — | 53.8% |
| FrontierCode | — | 42.4% |
| LMArena WebDev | — | 1552 |
| SciCode | — | 54.1% |
| GSO | 2.9% | — |
| WeirdML | — | 46.4% |
| ALE-Bench | — | 1,309 |
Agentic & Tool Use Not comparable
GLM-4.5-Air: —, Grok 4.5: 44.4 (#15)
| Benchmark | GLM-4.5-Air | Grok 4.5 |
|---|---|---|
| APEX-Agents | — | 56.2% |
| τ²-bench Banking | — | 47.9% |
| PostTrainBench | — | 23.4% |
| GBAEval | — | 65.4% |
| GDP.pdf | — | 14% |
| LMArena Search | — | 1213 |
| Vending-Bench 2 | — | 3,887 |
Reasoning Grok 4.5 leads
GLM-4.5-Air: 23.3 (#166), Grok 4.5: 56.0 (#23)
| Benchmark | GLM-4.5-Air | Grok 4.5 |
|---|---|---|
| Kagi LLM Benchmark | 43% | 83.5% |
| LMArena Hard Prompts | 1379 | 1462 |
| ARC-AGI-2 | — | 52.6% |
| SimpleBench | — | 70% |
| NYT Connections (extended) | — | 79.9% |
| ARC-AGI-1 | — | 87.2% |
| CritPt | — | 15.4% |
| Chess Puzzles | — | 36% |
| DTBench | — | 96.5% |
| LMCA | — | 45.2% |
| Surface Evolver Bench | — | 74.4% |
| Epoch Capabilities Index | — | 154.02 |
| ForecastBench | 59.2 | — |
Math Grok 4.5 leads
GLM-4.5-Air: 33.8 (#166), Grok 4.5: 60.7 (#29)
| Benchmark | GLM-4.5-Air | Grok 4.5 |
|---|---|---|
| LMArena Math | 1396 | 1457 |
| FrontierMath (Tiers 1-3) | — | 57.2% |
| FrontierMath Tier 4 | — | 24.4% |
| OTIS Mock AIME 2024-2025 | — | 97.8% |
| ProofBench | — | 31% |
| Omni-MATH | 39.1% | — |
Knowledge Grok 4.5 leads
GLM-4.5-Air: 34.6 (#184), Grok 4.5: 62.3 (#21)
| Benchmark | GLM-4.5-Air | Grok 4.5 |
|---|---|---|
| LMArena Expert | 1370 | 1466 |
| GPQA Diamond | — | 93.4% |
| Humanity's Last Exam | 8.1% | — |
| SimpleQA Verified | — | 48.3% |
| MMLU-Pro | 76.2% | — |
| Vectara Hallucination Rate | 9.3% | — |
| GPQA (HELM) | 59.4% | — |
Multimodal Not comparable
GLM-4.5-Air: —, Grok 4.5: 37.8 (#63)
| Benchmark | GLM-4.5-Air | Grok 4.5 |
|---|---|---|
| LMArena Vision | — | 1287 |
| Blueprint-Bench 2 | — | 27.3% |
| Furniture Assembly | — | 22.5% |
| LMArena Document | — | 1452 |
Multilingual Grok 4.5 leads
GLM-4.5-Air: 49.4 (#130), Grok 4.5: 54.5 (#43)
| Benchmark | GLM-4.5-Air | Grok 4.5 |
|---|---|---|
| LMArena Non-English | 1366 | 1439 |
| LMArena Chinese | 1427 | 1497 |
| LMArena French | 1399 | 1458 |
| LMArena German | 1376 | 1448 |
| LMArena Japanese | 1348 | 1425 |
| LMArena Korean | 1306 | 1402 |
| LMArena Russian | 1373 | 1447 |
| LMArena Spanish | 1385 | 1452 |
Instruction Following Grok 4.5 leads
GLM-4.5-Air: 69.8 (#165), Grok 4.5: 76.3 (#45)
| Benchmark | GLM-4.5-Air | Grok 4.5 |
|---|---|---|
| LMArena Instruction Following | 1353 | 1446 |
| IFEval | 81.2% | — |
Long Context Grok 4.5 leads
GLM-4.5-Air: 41.6 (#130), Grok 4.5: 44.8 (#54)
| Benchmark | GLM-4.5-Air | Grok 4.5 |
|---|---|---|
| LMArena Longer Query | 1367 | 1463 |
Writing & Preference Grok 4.5 leads
GLM-4.5-Air: 56.1 (#134), Grok 4.5: 65.9 (#40)
| Benchmark | GLM-4.5-Air | Grok 4.5 |
|---|---|---|
| LMArena Text | 1384 | 1448 |
| LMArena Creative Writing | 1343 | 1442 |
| LMArena Multi-Turn | 1371 | 1457 |
| EQ-Bench Creative Writing | — | 1579 |
| WildBench | 78.9% | — |
Frequently asked questions
Is GLM-4.5-Air better than Grok 4.5?
Grok 4.5 is the stronger model overall, scoring 55.0 to 38.7 on the Noometry Index. GLM-4.5-Air costs 7.1× less per token, which makes it the better buy when Grok 4.5's lead doesn't matter for your workload.
Which is cheaper, GLM-4.5-Air or Grok 4.5?
GLM-4.5-Air is cheaper. It lists at $0.20 per million input tokens and $1.10 per million output tokens; Grok 4.5 lists at $2 and $6.
Is GLM-4.5-Air or Grok 4.5 better for coding?
Grok 4.5 scores higher on coding benchmarks: 51.7 versus 34.5 in the Noometry coding category.
Which has the bigger context window?
Grok 4.5 does, with 500K tokens against 131K.
How many benchmarks do GLM-4.5-Air and Grok 4.5 share?
18 benchmarks have published results for both models. GLM-4.5-Air has 27 scored results on Noometry and Grok 4.5 has 52.