Model comparison
GLM-4.5-Air vs GPT-5 Mini
GPT-5 Mini is the stronger model overall, scoring 41.8 to 38.9 on the Noometry Index. GLM-4.5-Air costs 1.6× less per token, which makes it the better buy when GPT-5 Mini's lead doesn't matter for your workload.
Last verified . 26 shared benchmarks.
Summary
- They share 26 benchmarks with published results for both. GLM-4.5-Air scores higher in 3 categories and GPT-5 Mini in 5 categories; 4 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where GPT-5 Mini leads 45.6 to 35.0.
- The biggest single-benchmark swing is Omni-MATH: 39.1% for GLM-4.5-Air and 72.2% for GPT-5 Mini.
- GLM-4.5-Air is cheaper at $0.20 / $1.10 per million input/output tokens, against $0.25 / $2 for GPT-5 Mini.
- GPT-5 Mini accepts more context: 400K tokens versus 131K.
- GLM-4.5-Air has downloadable open weights; the other is API-only.
Side by side
| GLM-4.5-Air | GPT-5 Mini | |
|---|---|---|
| Provider | Z.ai (Zhipu) | OpenAI |
| Noometry Index | 38.9 | 41.8 |
| Released | 2025-07-20 | 2025-08-07 |
| Weights | Open | Proprietary |
| Context window | 131K | 400K |
| Max output | 98K | 128K |
| Input $ / M tokens | $0.20 | $0.25 |
| Output $ / M tokens | $1.10 | $2 |
| Results tracked | 27 | 60 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5 Mini leads
GLM-4.5-Air: 33.3 (#259), GPT-5 Mini: 40.1 (#146)
| Benchmark | GLM-4.5-Air | GPT-5 Mini |
|---|---|---|
| LMArena Coding | 1397 | 1406 |
| SWE-bench Verified | — | 64.7% |
| SWE-bench Verified (bash only) | — | 59.8% |
| SWE-bench Multilingual | — | 39.7% |
| SciCode | — | 39.2% |
| GSO | 2.9% | — |
| WeirdML | — | 52.7% |
| ALE-Bench | — | 799.77 |
| AlgoTune | — | 1.38 |
Agentic & Tool Use Not comparable
GLM-4.5-Air: —, GPT-5 Mini: 31.1 (#70)
| Benchmark | GLM-4.5-Air | GPT-5 Mini |
|---|---|---|
| Terminal-Bench | — | 34.8% |
| Berkeley Function Calling Leaderboard | — | 55.5% |
| Vending-Bench 2 | — | -31.18 |
Reasoning Too close to call
GLM-4.5-Air: 24.1 (#166), GPT-5 Mini: 23.9 (#168)
| Benchmark | GLM-4.5-Air | GPT-5 Mini |
|---|---|---|
| Kagi LLM Benchmark | 43% | 70.3% |
| LMArena Hard Prompts | 1379 | 1380 |
| ForecastBench | 59.2 | 61 |
| ARC-AGI-2 | — | 4.4% |
| ARC-AGI-1 | — | 54.3% |
| CritPt | — | 0% |
| Chess Puzzles | — | 30% |
| EnigmaEval | — | 8.2% |
| Mystery Game Puzzles | — | 10% |
| DTBench | — | 80.5% |
| LMCA | — | 34.2% |
| Epoch Capabilities Index | — | 145.52 |
Math GPT-5 Mini leads
GLM-4.5-Air: 36.2 (#170), GPT-5 Mini: 46.7 (#69)
| Benchmark | GLM-4.5-Air | GPT-5 Mini |
|---|---|---|
| Omni-MATH | 39.1% | 72.2% |
| LMArena Math | 1396 | 1378 |
| FrontierMath (Tiers 1-3) | — | 46.7% |
| FrontierMath Tier 4 | — | 12.2% |
| OTIS Mock AIME 2024-2025 | — | 86.7% |
| ProofBench | — | 9% |
| MATH Level 5 | — | 97.8% |
| FrontierMath (Feb 2025 set) | — | 27.2% |
| FrontierMath Tier 4 (v1) | — | 6.3% |
Knowledge GPT-5 Mini leads
GLM-4.5-Air: 35.0 (#191), GPT-5 Mini: 45.6 (#86)
| Benchmark | GLM-4.5-Air | GPT-5 Mini |
|---|---|---|
| Humanity's Last Exam | 8.1% | 19.4% |
| MMLU-Pro | 76.2% | 83.5% |
| Vectara Hallucination Rate | 9.3% | 12.9% |
| GPQA (HELM) | 59.4% | 75.6% |
| LMArena Expert | 1370 | 1379 |
| GPQA Diamond | — | 75% |
| SimpleQA Verified | — | 21.6% |
| Confabulations | — | 13.3% |
Multimodal Not comparable
GLM-4.5-Air: —, GPT-5 Mini: 35.6 (#85)
| Benchmark | GLM-4.5-Air | GPT-5 Mini |
|---|---|---|
| LMArena Vision | — | 1202 |
| VPCT | — | 40.2% |
Multilingual Too close to call
GLM-4.5-Air: 49.1 (#135), GPT-5 Mini: 48.9 (#137)
| Benchmark | GLM-4.5-Air | GPT-5 Mini |
|---|---|---|
| LMArena Non-English | 1366 | 1363 |
| LMArena Chinese | 1426 | 1385 |
| LMArena French | 1399 | 1386 |
| LMArena German | 1377 | 1366 |
| LMArena Japanese | 1348 | 1341 |
| LMArena Korean | 1308 | 1308 |
| LMArena Russian | 1373 | 1362 |
| LMArena Spanish | 1386 | 1355 |
Instruction Following GPT-5 Mini leads
GLM-4.5-Air: 69.6 (#171), GPT-5 Mini: 76.2 (#46)
| Benchmark | GLM-4.5-Air | GPT-5 Mini |
|---|---|---|
| IFEval | 81.2% | 92.7% |
| LMArena Instruction Following | 1354 | 1357 |
Long Context Too close to call
GLM-4.5-Air: 41.6 (#135), GPT-5 Mini: 41.9 (#132)
| Benchmark | GLM-4.5-Air | GPT-5 Mini |
|---|---|---|
| LMArena Longer Query | 1366 | 1355 |
| Fiction.LiveBench | — | 69.4% |
Writing & Preference Too close to call
GLM-4.5-Air: 55.9 (#139), GPT-5 Mini: 55.2 (#148)
| Benchmark | GLM-4.5-Air | GPT-5 Mini |
|---|---|---|
| LMArena Text | 1384 | 1373 |
| LMArena Creative Writing | 1343 | 1325 |
| WildBench | 78.9% | 85.5% |
| LMArena Multi-Turn | 1371 | 1363 |
| Short-Story Creative Writing | — | 83.1% |
| EQ-Bench Creative Writing | — | 1313 |
Frequently asked questions
Is GLM-4.5-Air better than GPT-5 Mini?
GPT-5 Mini is the stronger model overall, scoring 41.8 to 38.9 on the Noometry Index. GLM-4.5-Air costs 1.6× less per token, which makes it the better buy when GPT-5 Mini's lead doesn't matter for your workload.
Which is cheaper, GLM-4.5-Air or GPT-5 Mini?
GLM-4.5-Air is cheaper. It lists at $0.20 per million input tokens and $1.10 per million output tokens; GPT-5 Mini lists at $0.25 and $2.
Is GLM-4.5-Air or GPT-5 Mini better for coding?
GPT-5 Mini scores higher on coding benchmarks: 40.1 versus 33.3 in the Noometry coding category.
Which has the bigger context window?
GPT-5 Mini does, with 400K tokens against 131K.
How many benchmarks do GLM-4.5-Air and GPT-5 Mini share?
26 benchmarks have published results for both models. GLM-4.5-Air has 27 scored results on Noometry and GPT-5 Mini has 60.