Model comparison
GLM-5.2 vs GPT-5.4 Pro
GPT-5.4 Pro is the stronger model overall, scoring 58.9 to 51.1 on the Noometry Index. GLM-5.2 costs 31× less per token, which makes it the better buy when GPT-5.4 Pro's lead doesn't matter for your workload.
Last verified . 11 shared benchmarks.
Summary
- They share 11 benchmarks with published results for both. GLM-5.2 scores higher in 1 category and GPT-5.4 Pro in 3 categories; 4 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where GPT-5.4 Pro leads 70.7 to 42.3.
- The biggest single-benchmark swing is ARC-AGI-2: 22.8% for GLM-5.2 and 83.3% for GPT-5.4 Pro.
- GLM-5.2 is cheaper at $1.40 / $4.40 per million input/output tokens, against $30 / $180 for GPT-5.4 Pro.
- GPT-5.4 Pro accepts more context: 1.05M tokens versus 1M.
- GLM-5.2 has downloadable open weights; the other is API-only.
Side by side
| GLM-5.2 | GPT-5.4 Pro | |
|---|---|---|
| Provider | Z.ai (Zhipu) | OpenAI |
| Noometry Index | 51.1 | 58.9 |
| Released | 2026-06-13 | 2026-03-05 |
| Weights | Open | Proprietary |
| Context window | 1M | 1.05M |
| Max output | 131K | 128K |
| Input $ / M tokens | $1.40 | $30 |
| Output $ / M tokens | $4.40 | $180 |
| Results tracked | 51 | 16 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GLM-5.2 leads
GLM-5.2: 51.3 (#41), GPT-5.4 Pro: 43.2 (#90)
| Benchmark | GLM-5.2 | GPT-5.4 Pro |
|---|---|---|
| WeirdML | 70.1% | 57.4% |
| SWE-bench Verified | 78.7% | — |
| DeepSWE | 43.8% | — |
| FrontierCode | 24.5% | — |
| LMArena WebDev | 1603 | — |
| SciCode | 50.5% | — |
| LMArena Coding | 1485 | — |
| ALE-Bench | 1,047 | — |
Agentic & Tool Use Not comparable
GLM-5.2: 32.4 (#63), GPT-5.4 Pro: —
| Benchmark | GLM-5.2 | GPT-5.4 Pro |
|---|---|---|
| APEX-Agents | 45.2% | — |
| τ²-bench Banking | 37.1% | — |
| PostTrainBench | 31.7% | — |
| GBAEval | 0% | — |
| Vending-Bench 2 | 8,314 | — |
Reasoning GPT-5.4 Pro leads
GLM-5.2: 42.3 (#52), GPT-5.4 Pro: 70.7 (#13)
| Benchmark | GLM-5.2 | GPT-5.4 Pro |
|---|---|---|
| ARC-AGI-2 | 22.8% | 83.3% |
| SimpleBench | 58.8% | 74.1% |
| ARC-AGI-1 | 77% | 94.5% |
| CritPt | 20.9% | 30% |
| Chess Puzzles | 21% | 58.6% |
| Epoch Capabilities Index | 151.78 | 158.93 |
| Kagi LLM Benchmark | 62.6% | — |
| NYT Connections (extended) | 74.3% | — |
| EnigmaEval | — | 23.8% |
| EBR-Bench | 9.5% | — |
| LMArena Hard Prompts | 1480 | — |
| Mystery Game Puzzles | 19% | — |
| DTBench | 93.6% | — |
| LMCA | 45.8% | — |
| Surface Evolver Bench | 55.6% | — |
Math GPT-5.4 Pro leads
GLM-5.2: 55.7 (#43), GPT-5.4 Pro: 72.4 (#22)
| Benchmark | GLM-5.2 | GPT-5.4 Pro |
|---|---|---|
| FrontierMath (Tiers 1-3) | 59.2% | 82.5% |
| FrontierMath Tier 4 | 29.3% | 58.5% |
| MathArena Final-Answer Competitions | 67.6% | — |
| OTIS Mock AIME 2024-2025 | 86.4% | — |
| ProofBench | 35% | — |
| LMArena Math | 1482 | — |
| FrontierMath (Feb 2025 set) | — | 50% |
| FrontierMath Tier 4 (v1) | — | 37.5% |
Knowledge GPT-5.4 Pro leads
GLM-5.2: 57.1 (#40), GPT-5.4 Pro: 68.3 (#7)
| Benchmark | GLM-5.2 | GPT-5.4 Pro |
|---|---|---|
| GPQA Diamond | 91.9% | 94.6% |
| SimpleQA Verified | 34.2% | 46.3% |
| Humanity's Last Exam | — | 44.3% |
| Vectara Hallucination Rate | — | 8.3% |
| LMArena Expert | 1486 | — |
Multilingual Not comparable
GLM-5.2: 55.8 (#26), GPT-5.4 Pro: —
| Benchmark | GLM-5.2 | GPT-5.4 Pro |
|---|---|---|
| LMArena Non-English | 1459 | — |
| LMArena Chinese | 1519 | — |
| LMArena French | 1479 | — |
| LMArena German | 1468 | — |
| LMArena Japanese | 1451 | — |
| LMArena Korean | 1445 | — |
| LMArena Russian | 1466 | — |
| LMArena Spanish | 1477 | — |
Instruction Following Not comparable
GLM-5.2: 76.9 (#34), GPT-5.4 Pro: —
| Benchmark | GLM-5.2 | GPT-5.4 Pro |
|---|---|---|
| LMArena Instruction Following | 1465 | — |
Long Context Not comparable
GLM-5.2: 45.3 (#43), GPT-5.4 Pro: —
| Benchmark | GLM-5.2 | GPT-5.4 Pro |
|---|---|---|
| LMArena Longer Query | 1479 | — |
Writing & Preference Not comparable
GLM-5.2: 70.4 (#21), GPT-5.4 Pro: —
| Benchmark | GLM-5.2 | GPT-5.4 Pro |
|---|---|---|
| LMArena Text | 1470 | — |
| LMArena Creative Writing | 1462 | — |
| EQ-Bench Creative Writing | 1757 | — |
| EQ-Bench 4 | 1222 | — |
| LMArena Multi-Turn | 1469 | — |
Frequently asked questions
Is GLM-5.2 better than GPT-5.4 Pro?
GPT-5.4 Pro is the stronger model overall, scoring 58.9 to 51.1 on the Noometry Index. GLM-5.2 costs 31× less per token, which makes it the better buy when GPT-5.4 Pro's lead doesn't matter for your workload.
Which is cheaper, GLM-5.2 or GPT-5.4 Pro?
GLM-5.2 is cheaper. It lists at $1.40 per million input tokens and $4.40 per million output tokens; GPT-5.4 Pro lists at $30 and $180.
Is GLM-5.2 or GPT-5.4 Pro better for coding?
GLM-5.2 scores higher on coding benchmarks: 51.3 versus 43.2 in the Noometry coding category.
Which has the bigger context window?
GPT-5.4 Pro does, with 1.05M tokens against 1M.
How many benchmarks do GLM-5.2 and GPT-5.4 Pro share?
11 benchmarks have published results for both models. GLM-5.2 has 51 scored results on Noometry and GPT-5.4 Pro has 16.