Model comparison
GLM-4.5-Air vs o1
o1 is the stronger model overall, scoring 40.9 to 38.9 on the Noometry Index. GLM-4.5-Air costs 62× less per token, which makes it the better buy when o1'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 3 categories and o1 in 5 categories; 5 gaps are clear of the uncertainty.
- The widest gap is in coding, where o1 leads 46.1 to 33.3.
- GLM-4.5-Air is cheaper at $0.20 / $1.10 per million input/output tokens, against $15 / $60 for o1.
- o1 accepts more context: 200K tokens versus 131K.
- GLM-4.5-Air has downloadable open weights; the other is API-only.
Side by side
| GLM-4.5-Air | o1 | |
|---|---|---|
| Provider | Z.ai (Zhipu) | OpenAI |
| Noometry Index | 38.9 | 40.9 |
| Released | 2025-07-20 | 2024-09-12 |
| Weights | Open | Proprietary |
| Context window | 131K | 200K |
| Max output | 98K | 100K |
| Input $ / M tokens | $0.20 | $15 |
| Output $ / M tokens | $1.10 | $60 |
| Results tracked | 27 | 52 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding o1 leads
GLM-4.5-Air: 33.3 (#259), o1: 46.1 (#70)
| Benchmark | GLM-4.5-Air | o1 |
|---|---|---|
| LMArena Coding | 1397 | 1367 |
| Aider Polyglot | — | 61.7% |
| GSO | 2.9% | — |
| WeirdML | — | 47.6% |
| LiveBench Coding | — | 69.7% |
| CadEval | — | 56% |
| HumanEval+ | — | 89% |
| MBPP+ | — | 80.2% |
Agentic & Tool Use Not comparable
GLM-4.5-Air: —, o1: 24.6 (#117)
| Benchmark | GLM-4.5-Air | o1 |
|---|---|---|
| Cybench | — | 10% |
| METR Time Horizons | — | 51.1% |
Reasoning o1 leads
GLM-4.5-Air: 24.1 (#166), o1: 27.9 (#111)
| Benchmark | GLM-4.5-Air | o1 |
|---|---|---|
| LMArena Hard Prompts | 1379 | 1371 |
| SimpleBench | — | 41.7% |
| Kagi LLM Benchmark | 43% | — |
| ARC-AGI-1 | — | 30.7% |
| Chess Puzzles | — | 15% |
| EnigmaEval | — | 5.7% |
| LiveBench Reasoning | — | 91.6% |
| DTBench | — | 74.7% |
| LiveBench Data Analysis | — | 65.5% |
| LMCA | — | 22.3% |
| Epoch Capabilities Index | — | 141.91 |
| ForecastBench | 59.2 | — |
| LiveBench | — | 75.7% |
Math Too close to call
GLM-4.5-Air: 36.2 (#170), o1: 36.1 (#175)
| Benchmark | GLM-4.5-Air | o1 |
|---|---|---|
| LMArena Math | 1396 | 1388 |
| FrontierMath (Tiers 1-3) | — | 14.7% |
| OTIS Mock AIME 2024-2025 | — | 73.3% |
| Omni-MATH | 39.1% | — |
| LiveBench Math | — | 80.3% |
| MATH Level 5 | — | 94.7% |
| FrontierMath (Feb 2025 set) | — | 9.3% |
Knowledge o1 leads
GLM-4.5-Air: 35.0 (#191), o1: 41.5 (#110)
| Benchmark | GLM-4.5-Air | o1 |
|---|---|---|
| Humanity's Last Exam | 8.1% | 8% |
| LMArena Expert | 1370 | 1361 |
| GPQA Diamond | — | 76.8% |
| SimpleQA Verified | — | 41.1% |
| MMLU-Pro | 76.2% | — |
| Confabulations | — | 11.7% |
| Vectara Hallucination Rate | 9.3% | — |
| GPQA (HELM) | 59.4% | — |
Multimodal Not comparable
GLM-4.5-Air: —, o1: 34.2 (#93)
| Benchmark | GLM-4.5-Air | o1 |
|---|---|---|
| LMArena Vision | — | 1168 |
| GeoBench | — | 80% |
| VPCT | — | 37% |
| SpatialViz-Bench | — | 41.4% |
Multilingual Too close to call
GLM-4.5-Air: 49.1 (#135), o1: 48.6 (#142)
| Benchmark | GLM-4.5-Air | o1 |
|---|---|---|
| LMArena Non-English | 1366 | 1358 |
| LMArena Chinese | 1426 | 1394 |
| LMArena French | 1399 | 1344 |
| LMArena German | 1377 | 1337 |
| LMArena Japanese | 1348 | 1346 |
| LMArena Korean | 1308 | 1396 |
| LMArena Russian | 1373 | 1356 |
| LMArena Spanish | 1386 | 1345 |
Instruction Following o1 leads
GLM-4.5-Air: 69.6 (#171), o1: 74.8 (#86)
| Benchmark | GLM-4.5-Air | o1 |
|---|---|---|
| LMArena Instruction Following | 1354 | 1367 |
| LiveBench Instruction Following | — | 81.5% |
| IFEval | 81.2% | — |
Long Context o1 leads
GLM-4.5-Air: 41.6 (#135), o1: 50.3 (#9)
| Benchmark | GLM-4.5-Air | o1 |
|---|---|---|
| LMArena Longer Query | 1366 | 1378 |
| Fiction.LiveBench | — | 83.3% |
Writing & Preference Too close to call
GLM-4.5-Air: 55.9 (#139), o1: 55.6 (#144)
| Benchmark | GLM-4.5-Air | o1 |
|---|---|---|
| LMArena Text | 1384 | 1366 |
| LMArena Creative Writing | 1343 | 1348 |
| LMArena Multi-Turn | 1371 | 1369 |
| Short-Story Creative Writing | — | 70.2% |
| WildBench | 78.9% | — |
| LiveBench Language | — | 65.4% |
Frequently asked questions
Is GLM-4.5-Air better than o1?
o1 is the stronger model overall, scoring 40.9 to 38.9 on the Noometry Index. GLM-4.5-Air costs 62× less per token, which makes it the better buy when o1's lead doesn't matter for your workload.
Which is cheaper, GLM-4.5-Air or o1?
GLM-4.5-Air is cheaper. It lists at $0.20 per million input tokens and $1.10 per million output tokens; o1 lists at $15 and $60.
Is GLM-4.5-Air or o1 better for coding?
o1 scores higher on coding benchmarks: 46.1 versus 33.3 in the Noometry coding category.
Which has the bigger context window?
o1 does, with 200K tokens against 131K.
How many benchmarks do GLM-4.5-Air and o1 share?
18 benchmarks have published results for both models. GLM-4.5-Air has 27 scored results on Noometry and o1 has 52.