Model comparison
GLM-5.2 vs GPT-5.1
GLM-5.2 is the stronger model overall, scoring 50.9 to 48.8 on the Noometry Index.
Last verified . 35 shared benchmarks.
Summary
- They share 35 benchmarks with published results for both. GLM-5.2 scores higher in 6 categories and GPT-5.1 in 3 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in instruction following, where GPT-5.1 leads 84.0 to 77.1.
- The biggest single-benchmark swing is CritPt: 20.9% for GLM-5.2 and 4.9% for GPT-5.1.
- GLM-5.2 is cheaper at $1.40 / $4.40 per million input/output tokens, against $1.25 / $10 for GPT-5.1.
- 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-5.1 | |
|---|---|---|
| Provider | Z.ai (Zhipu) | OpenAI |
| Noometry Index | 50.9 | 48.8 |
| Released | 2026-06-13 | 2025-11-13 |
| Weights | Open | Proprietary |
| Context window | 1M | 400K |
| Max output | 131K | 128K |
| Input $ / M tokens | $1.40 | $1.25 |
| Output $ / M tokens | $4.40 | $10 |
| Results tracked | 50 | 63 |
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-5.1: 46.9 (#59)
| Benchmark | GLM-5.2 | GPT-5.1 |
|---|---|---|
| SWE-bench Verified | 78.7% | 68% |
| LMArena WebDev | 1604 | 1395 |
| SciCode | 50.5% | 43.3% |
| WeirdML | 70.1% | 60.8% |
| LMArena Coding | 1484 | 1454 |
| ALE-Bench | 1,047 | 1,192 |
| DeepSWE | 43.8% | — |
| FrontierCode | 24.5% | — |
| SWE-bench Verified (bash only) | — | 66% |
| GSO | — | 13.7% |
| LiveBench Coding | — | 72.5% |
Agentic & Tool Use Too close to call
GLM-5.2: 31.6 (#63), GPT-5.1: 32.5 (#55)
| Benchmark | GLM-5.2 | GPT-5.1 |
|---|---|---|
| Vending-Bench 2 | 8,314 | 1,473 |
| Terminal-Bench | — | 47.6% |
| τ²-bench Banking | 37.1% | — |
| DeepResearch Bench | — | 42.8% |
| PostTrainBench | 31.7% | — |
| GBAEval | 0% | — |
| LMArena Search | — | 1199 |
Reasoning GLM-5.2 leads
GLM-5.2: 42.1 (#46), GPT-5.1: 39.9 (#50)
| Benchmark | GLM-5.2 | GPT-5.1 |
|---|---|---|
| ARC-AGI-2 | 22.8% | 17.6% |
| SimpleBench | 58.8% | 53.2% |
| ARC-AGI-1 | 77% | 72.8% |
| CritPt | 20.9% | 4.9% |
| Chess Puzzles | 21% | 32% |
| LMArena Hard Prompts | 1480 | 1457 |
| Mystery Game Puzzles | 19% | 19% |
| DTBench | 93.6% | 90.1% |
| LMCA | 45.8% | 43.9% |
| Epoch Capabilities Index | 151.76 | 149.66 |
| Kagi LLM Benchmark | 62.6% | — |
| NYT Connections (extended) | 74.3% | — |
| EnigmaEval | — | 11.2% |
| EBR-Bench | 9.5% | — |
| LiveBench Reasoning | — | 95.8% |
| LiveBench Data Analysis | — | 72.1% |
| Surface Evolver Bench | 55.6% | — |
| ForecastBench | — | 58.1 |
| LiveBench | — | 78.8% |
Math GLM-5.2 leads
GLM-5.2: 55.4 (#35), GPT-5.1: 50.3 (#45)
| Benchmark | GLM-5.2 | GPT-5.1 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 86.4% | 88.6% |
| LMArena Math | 1482 | 1446 |
| FrontierMath (Tiers 1-3) | 59.2% | — |
| FrontierMath Tier 4 | 29.3% | — |
| MathArena Final-Answer Competitions | 67.6% | — |
| ProofBench | 35% | — |
| Omni-MATH | — | 46.4% |
| LiveBench Math | — | 94.5% |
| FrontierMath (Feb 2025 set) | — | 31% |
| FrontierMath Tier 4 (v1) | — | 12.5% |
Knowledge GLM-5.2 leads
GLM-5.2: 57.1 (#36), GPT-5.1: 50.5 (#65)
| Benchmark | GLM-5.2 | GPT-5.1 |
|---|---|---|
| GPQA Diamond | 91.9% | 87.6% |
| SimpleQA Verified | 34.2% | 48% |
| LMArena Expert | 1484 | 1470 |
| Humanity's Last Exam | — | 23.7% |
| MMLU-Pro | — | 57.9% |
| Vectara Hallucination Rate | — | 10.9% |
| GPQA (HELM) | — | 44.2% |
Multimodal Not comparable
GLM-5.2: —, GPT-5.1: 44.0 (#17)
| Benchmark | GLM-5.2 | GPT-5.1 |
|---|---|---|
| LMArena Vision | — | 1250 |
| VPCT | — | 58.7% |
| LMArena Document | — | 1403 |
Multilingual GLM-5.2 leads
GLM-5.2: 56.0 (#25), GPT-5.1: 54.0 (#51)
| Benchmark | GLM-5.2 | GPT-5.1 |
|---|---|---|
| LMArena Non-English | 1459 | 1431 |
| LMArena Chinese | 1518 | 1495 |
| LMArena French | 1476 | 1451 |
| LMArena German | 1470 | 1439 |
| LMArena Japanese | 1451 | 1453 |
| LMArena Korean | 1445 | 1401 |
| LMArena Russian | 1467 | 1435 |
| LMArena Spanish | 1477 | 1430 |
Instruction Following GPT-5.1 leads
GLM-5.2: 77.1 (#32), GPT-5.1: 84.0 (#1)
| Benchmark | GLM-5.2 | GPT-5.1 |
|---|---|---|
| LMArena Instruction Following | 1465 | 1442 |
| LiveBench Instruction Following | — | 93.3% |
| IFEval | — | 93.5% |
Long Context GPT-5.1 leads
GLM-5.2: 45.3 (#43), GPT-5.1: 47.6 (#14)
| Benchmark | GLM-5.2 | GPT-5.1 |
|---|---|---|
| LMArena Longer Query | 1479 | 1448 |
| CL-bench | — | 23.7% |
| CL-bench Life | — | 17.3% |
Writing & Preference GLM-5.2 leads
GLM-5.2: 70.7 (#20), GPT-5.1: 64.7 (#53)
| Benchmark | GLM-5.2 | GPT-5.1 |
|---|---|---|
| LMArena Text | 1470 | 1443 |
| LMArena Creative Writing | 1465 | 1427 |
| LMArena Multi-Turn | 1468 | 1449 |
| EQ-Bench Creative Writing | 1757 | — |
| WildBench | — | 86.3% |
| EQ-Bench 4 | 1222 | — |
| LiveBench Language | — | 80.2% |
Frequently asked questions
Is GLM-5.2 better than GPT-5.1?
GLM-5.2 is the stronger model overall, scoring 50.9 to 48.8 on the Noometry Index.
Which is cheaper, GLM-5.2 or GPT-5.1?
GLM-5.2 is cheaper. It lists at $1.40 per million input tokens and $4.40 per million output tokens; GPT-5.1 lists at $1.25 and $10.
Is GLM-5.2 or GPT-5.1 better for coding?
GLM-5.2 scores higher on coding benchmarks: 50.9 versus 46.9 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-5.1 share?
35 benchmarks have published results for both models. GLM-5.2 has 50 scored results on Noometry and GPT-5.1 has 63.