Model comparison
GLM-5 vs GPT-4o
GLM-5 is the stronger model overall, scoring 46.1 to 28.6 on the Noometry Index.
Last verified . 30 shared benchmarks.
Summary
- They share 30 benchmarks with published results for both. GLM-5 scores higher in 9 categories and GPT-4o in 0 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where GLM-5 leads 46.4 to 10.6.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 80% for GLM-5 and 6.4% for GPT-4o.
- GLM-5 is cheaper at $1 / $3.20 per million input/output tokens, against $2.50 / $10 for GPT-4o.
- GLM-5 accepts more context: 205K tokens versus 128K.
- GLM-5 has downloadable open weights; the other is API-only.
Side by side
| GLM-5 | GPT-4o | |
|---|---|---|
| Provider | Z.ai (Zhipu) | OpenAI |
| Noometry Index | 46.1 | 28.6 |
| Released | 2026-02-11 | 2024-05-13 |
| Weights | Open | Proprietary |
| Context window | 205K | 128K |
| Max output | 131K | 16K |
| Input $ / M tokens | $1 | $2.50 |
| Output $ / M tokens | $3.20 | $10 |
| Results tracked | 45 | 72 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GLM-5 leads
GLM-5: 49.0 (#52), GPT-4o: 24.8 (#328)
| Benchmark | GLM-5 | GPT-4o |
|---|---|---|
| SWE-bench Verified | 72.1% | 31% |
| SWE-bench Verified (bash only) | 72.8% | 21.6% |
| WeirdML | 48.2% | 25.1% |
| LMArena Coding | 1461 | 1297 |
| Aider Polyglot | — | 45.3% |
| LMArena WebDev | 1434 | — |
| SWE-bench Multilingual | 69.7% | — |
| GSO | — | 0% |
| BigCodeBench Instruct | — | 51.1% |
| LiveBench Coding | — | 51.4% |
| BigCodeBench Complete | — | 61.1% |
| CadEval | — | 26% |
| ALE-Bench | 765.62 | — |
| HumanEval+ | — | 87.2% |
| MBPP+ | — | 72.2% |
Agentic & Tool Use GLM-5 leads
GLM-5: 31.1 (#71), GPT-4o: 21.0 (#141)
| Benchmark | GLM-5 | GPT-4o |
|---|---|---|
| Terminal-Bench | 52.4% | — |
| GDPval | — | 9.9% |
| TheAgentCompany | — | 8.6% |
| τ²-bench Airline | 82.5% | — |
| τ²-bench Banking | 9.8% | — |
| τ²-bench Retail | 73.7% | — |
| τ²-bench Telecom | 86.8% | — |
| Cybench | — | 12.5% |
| BALROG | — | 32.3% |
| LMArena Search | — | 1006 |
| METR Time Horizons | — | 40.8% |
| Vending-Bench 2 | 4,432 | — |
Reasoning GLM-5 leads
GLM-5: 27.6 (#116), GPT-4o: 9.4 (#343)
| Benchmark | GLM-5 | GPT-4o |
|---|---|---|
| ARC-AGI-2 | 4.9% | 0% |
| SimpleBench | 53.2% | 17.8% |
| ARC-AGI-1 | 44.7% | 4.5% |
| Chess Puzzles | 10% | 13% |
| LMArena Hard Prompts | 1452 | 1281 |
| Epoch Capabilities Index | 145.83 | 128.97 |
| ForecastBench | 61 | 57.7 |
| Kagi LLM Benchmark | 75% | — |
| NYT Connections (extended) | 74.8% | — |
| CritPt | — | 0% |
| EnigmaEval | — | 0.8% |
| LiveBench Reasoning | — | 55.8% |
| DTBench | — | 64.5% |
| LiveBench Data Analysis | — | 60.9% |
| LMCA | — | 16.6% |
| LiveBench | — | 55.3% |
Math GLM-5 leads
GLM-5: 46.4 (#71), GPT-4o: 10.6 (#312)
| Benchmark | GLM-5 | GPT-4o |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 80% | 6.4% |
| LMArena Math | 1440 | 1285 |
| FrontierMath (Feb 2025 set) | 16.4% | 0.3% |
| FrontierMath (Tiers 1-3) | — | 0.4% |
| MathArena Final-Answer Competitions | 65.7% | — |
| Omni-MATH | — | 29.3% |
| LiveBench Math | — | 49.5% |
| MATH Level 5 | — | 53.3% |
| FrontierMath Tier 4 (v1) | 2.1% | — |
Knowledge GLM-5 leads
GLM-5: 52.3 (#64), GPT-4o: 28.8 (#242)
| Benchmark | GLM-5 | GPT-4o |
|---|---|---|
| GPQA Diamond | 87.8% | 49.2% |
| Vectara Hallucination Rate | 10.1% | 9.6% |
| LMArena Expert | 1454 | 1250 |
| Humanity's Last Exam | — | 2.7% |
| SimpleQA Verified | — | 26% |
| MMLU-Pro | — | 71.3% |
| Confabulations | — | 15.3% |
| GPQA (HELM) | — | 52% |
| MMLU | — | 88.1% |
Multimodal Not comparable
GLM-5: —, GPT-4o: 34.5 (#91)
| Benchmark | GLM-5 | GPT-4o |
|---|---|---|
| LMArena Vision | — | 1137 |
| Video-MME | — | 71.9% |
| GeoBench | — | 71% |
| VPCT | — | 40% |
| ScienceQA | — | 88.5% |
Multilingual GLM-5 leads
GLM-5: 53.7 (#58), GPT-4o: 43.2 (#186)
| Benchmark | GLM-5 | GPT-4o |
|---|---|---|
| LMArena Non-English | 1430 | 1283 |
| LMArena Chinese | 1511 | 1277 |
| LMArena French | 1455 | 1304 |
| LMArena German | 1445 | 1282 |
| LMArena Japanese | 1416 | 1257 |
| LMArena Korean | 1423 | 1234 |
| LMArena Russian | 1436 | 1286 |
| LMArena Spanish | 1454 | 1292 |
Instruction Following GLM-5 leads
GLM-5: 75.2 (#67), GPT-4o: 66.6 (#207)
| Benchmark | GLM-5 | GPT-4o |
|---|---|---|
| LMArena Instruction Following | 1428 | 1278 |
| LiveBench Instruction Following | — | 68.6% |
| IFEval | — | 81.7% |
Long Context GLM-5 leads
GLM-5: 44.7 (#60), GPT-4o: 39.4 (#179)
| Benchmark | GLM-5 | GPT-4o |
|---|---|---|
| LMArena Longer Query | 1446 | 1289 |
| Fiction.LiveBench | — | 66.7% |
| CL-bench | 18.7% | — |
Writing & Preference GLM-5 leads
GLM-5: 66.0 (#38), GPT-4o: 52.6 (#166)
| Benchmark | GLM-5 | GPT-4o |
|---|---|---|
| LMArena Text | 1446 | 1300 |
| LMArena Creative Writing | 1439 | 1292 |
| LMArena Multi-Turn | 1456 | 1302 |
| Short-Story Creative Writing | — | 81.8% |
| EQ-Bench Creative Writing | 1601 | — |
| WildBench | — | 82.8% |
| LiveBench Language | — | 47.6% |
Frequently asked questions
Is GLM-5 better than GPT-4o?
GLM-5 is the stronger model overall, scoring 46.1 to 28.6 on the Noometry Index.
Which is cheaper, GLM-5 or GPT-4o?
GLM-5 is cheaper. It lists at $1 per million input tokens and $3.20 per million output tokens; GPT-4o lists at $2.50 and $10.
Is GLM-5 or GPT-4o better for coding?
GLM-5 scores higher on coding benchmarks: 49.0 versus 24.8 in the Noometry coding category.
Which has the bigger context window?
GLM-5 does, with 205K tokens against 128K.
How many benchmarks do GLM-5 and GPT-4o share?
30 benchmarks have published results for both models. GLM-5 has 45 scored results on Noometry and GPT-4o has 72.