Model comparison
GLM-5.2 vs GPT-5 Mini
GLM-5.2 is the stronger model overall, scoring 50.9 to 41.7 on the Noometry Index. GPT-5 Mini costs 3.1× less per token, which makes it the better buy when GLM-5.2's lead doesn't matter for your workload.
Last verified . 38 shared benchmarks.
Summary
- They share 38 benchmarks with published results for both. GLM-5.2 scores higher in 9 categories and GPT-5 Mini in 0 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where GLM-5.2 leads 42.1 to 24.1.
- The biggest single-benchmark swing is ProofBench: 35% for GLM-5.2 and 9% for GPT-5 Mini.
- GPT-5 Mini is cheaper at $0.25 / $2 per million input/output tokens, against $1.40 / $4.40 for GLM-5.2.
- GLM-5.2 accepts more context: 1M tokens versus 400K.
- GLM-5.2 has downloadable open weights; the other is API-only.
Side by side
| GLM-5.2 | GPT-5 Mini | |
|---|---|---|
| Provider | Z.ai (Zhipu) | OpenAI |
| Noometry Index | 50.9 | 41.7 |
| Released | 2026-06-13 | 2025-08-07 |
| Weights | Open | Proprietary |
| Context window | 1M | 400K |
| Max output | 131K | 128K |
| Input $ / M tokens | $1.40 | $0.25 |
| Output $ / M tokens | $4.40 | $2 |
| Results tracked | 50 | 60 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GLM-5.2 leads
GLM-5.2: 50.9 (#37), GPT-5 Mini: 40.0 (#134)
| Benchmark | GLM-5.2 | GPT-5 Mini |
|---|---|---|
| SWE-bench Verified | 78.7% | 64.7% |
| SciCode | 50.5% | 39.2% |
| WeirdML | 70.1% | 52.7% |
| LMArena Coding | 1484 | 1406 |
| ALE-Bench | 1,047 | 799.77 |
| DeepSWE | 43.8% | — |
| FrontierCode | 24.5% | — |
| SWE-bench Verified (bash only) | — | 59.8% |
| LMArena WebDev | 1604 | — |
| SWE-bench Multilingual | — | 39.7% |
| AlgoTune | — | 1.38 |
Agentic & Tool Use Too close to call
GLM-5.2: 31.6 (#63), GPT-5 Mini: 31.0 (#64)
| Benchmark | GLM-5.2 | GPT-5 Mini |
|---|---|---|
| Vending-Bench 2 | 8,314 | -31.18 |
| Terminal-Bench | — | 34.8% |
| Berkeley Function Calling Leaderboard | — | 55.5% |
| τ²-bench Banking | 37.1% | — |
| PostTrainBench | 31.7% | — |
| GBAEval | 0% | — |
Reasoning GLM-5.2 leads
GLM-5.2: 42.1 (#46), GPT-5 Mini: 24.1 (#153)
| Benchmark | GLM-5.2 | GPT-5 Mini |
|---|---|---|
| ARC-AGI-2 | 22.8% | 4.4% |
| Kagi LLM Benchmark | 62.6% | 70.3% |
| ARC-AGI-1 | 77% | 54.3% |
| CritPt | 20.9% | 0% |
| Chess Puzzles | 21% | 30% |
| LMArena Hard Prompts | 1480 | 1380 |
| Mystery Game Puzzles | 19% | 10% |
| DTBench | 93.6% | 80.5% |
| LMCA | 45.8% | 34.2% |
| Epoch Capabilities Index | 151.76 | 145.51 |
| SimpleBench | 58.8% | — |
| NYT Connections (extended) | 74.3% | — |
| EnigmaEval | — | 8.2% |
| EBR-Bench | 9.5% | — |
| Surface Evolver Bench | 55.6% | — |
| ForecastBench | — | 61 |
Math GLM-5.2 leads
GLM-5.2: 55.4 (#35), GPT-5 Mini: 46.3 (#59)
| Benchmark | GLM-5.2 | GPT-5 Mini |
|---|---|---|
| FrontierMath (Tiers 1-3) | 59.2% | 46.7% |
| FrontierMath Tier 4 | 29.3% | 12.2% |
| OTIS Mock AIME 2024-2025 | 86.4% | 86.7% |
| ProofBench | 35% | 9% |
| LMArena Math | 1482 | 1377 |
| MathArena Final-Answer Competitions | 67.6% | — |
| Omni-MATH | — | 72.2% |
| MATH Level 5 | — | 97.8% |
| FrontierMath (Feb 2025 set) | — | 27.2% |
| FrontierMath Tier 4 (v1) | — | 6.3% |
Knowledge GLM-5.2 leads
GLM-5.2: 57.1 (#36), GPT-5 Mini: 45.5 (#80)
| Benchmark | GLM-5.2 | GPT-5 Mini |
|---|---|---|
| GPQA Diamond | 91.9% | 75% |
| SimpleQA Verified | 34.2% | 21.6% |
| LMArena Expert | 1484 | 1377 |
| Humanity's Last Exam | — | 19.4% |
| MMLU-Pro | — | 83.5% |
| Confabulations | — | 13.3% |
| Vectara Hallucination Rate | — | 12.9% |
| GPQA (HELM) | — | 75.6% |
Multimodal Not comparable
GLM-5.2: —, GPT-5 Mini: 35.0 (#80)
| Benchmark | GLM-5.2 | GPT-5 Mini |
|---|---|---|
| LMArena Vision | — | 1202 |
| VPCT | — | 40.2% |
Multilingual GLM-5.2 leads
GLM-5.2: 56.0 (#25), GPT-5 Mini: 49.1 (#131)
| Benchmark | GLM-5.2 | GPT-5 Mini |
|---|---|---|
| LMArena Non-English | 1459 | 1363 |
| LMArena Chinese | 1518 | 1387 |
| LMArena French | 1476 | 1385 |
| LMArena German | 1470 | 1366 |
| LMArena Japanese | 1451 | 1342 |
| LMArena Korean | 1445 | 1306 |
| LMArena Russian | 1467 | 1362 |
| LMArena Spanish | 1477 | 1355 |
Instruction Following Too close to call
GLM-5.2: 77.1 (#32), GPT-5 Mini: 76.4 (#43)
| Benchmark | GLM-5.2 | GPT-5 Mini |
|---|---|---|
| LMArena Instruction Following | 1465 | 1357 |
| IFEval | — | 92.7% |
Long Context GLM-5.2 leads
GLM-5.2: 45.3 (#43), GPT-5 Mini: 41.9 (#127)
| Benchmark | GLM-5.2 | GPT-5 Mini |
|---|---|---|
| LMArena Longer Query | 1479 | 1355 |
| Fiction.LiveBench | — | 69.4% |
Writing & Preference GLM-5.2 leads
GLM-5.2: 70.7 (#20), GPT-5 Mini: 55.4 (#143)
| Benchmark | GLM-5.2 | GPT-5 Mini |
|---|---|---|
| LMArena Text | 1470 | 1373 |
| LMArena Creative Writing | 1465 | 1325 |
| EQ-Bench Creative Writing | 1757 | 1313 |
| LMArena Multi-Turn | 1468 | 1363 |
| Short-Story Creative Writing | — | 83.1% |
| WildBench | — | 85.5% |
| EQ-Bench 4 | 1222 | — |
Frequently asked questions
Is GLM-5.2 better than GPT-5 Mini?
GLM-5.2 is the stronger model overall, scoring 50.9 to 41.7 on the Noometry Index. GPT-5 Mini costs 3.1× less per token, which makes it the better buy when GLM-5.2's lead doesn't matter for your workload.
Which is cheaper, GLM-5.2 or GPT-5 Mini?
GPT-5 Mini is cheaper. It lists at $0.25 per million input tokens and $2 per million output tokens; GLM-5.2 lists at $1.40 and $4.40.
Is GLM-5.2 or GPT-5 Mini better for coding?
GLM-5.2 scores higher on coding benchmarks: 50.9 versus 40.0 in the Noometry coding category.
Which has the bigger context window?
GLM-5.2 does, with 1M tokens against 400K.
How many benchmarks do GLM-5.2 and GPT-5 Mini share?
38 benchmarks have published results for both models. GLM-5.2 has 50 scored results on Noometry and GPT-5 Mini has 60.