Model comparison
GLM-5 vs GPT-4.1
GLM-5 is the stronger model overall, scoring 46.1 to 35.9 on the Noometry Index.
Last verified . 34 shared benchmarks.
Summary
- They share 34 benchmarks with published results for both. GLM-5 scores higher in 8 categories and GPT-4.1 in 1 category; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where GLM-5 leads 46.4 to 22.3.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 80% for GLM-5 and 38.3% for GPT-4.1.
- GLM-5 is cheaper at $1 / $3.20 per million input/output tokens, against $2 / $8 for GPT-4.1.
- GPT-4.1 accepts more context: 1.05M tokens versus 205K.
- GLM-5 has downloadable open weights; the other is API-only.
Side by side
| GLM-5 | GPT-4.1 | |
|---|---|---|
| Provider | Z.ai (Zhipu) | OpenAI |
| Noometry Index | 46.1 | 35.9 |
| Released | 2026-02-11 | 2025-04-14 |
| Weights | Open | Proprietary |
| Context window | 205K | 1.05M |
| Max output | 131K | 33K |
| Input $ / M tokens | $1 | $2 |
| Output $ / M tokens | $3.20 | $8 |
| Results tracked | 45 | 52 |
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-4.1: 34.4 (#238)
| Benchmark | GLM-5 | GPT-4.1 |
|---|---|---|
| SWE-bench Verified | 72.1% | 48.5% |
| SWE-bench Verified (bash only) | 72.8% | 39.6% |
| WeirdML | 48.2% | 39% |
| LMArena Coding | 1461 | 1391 |
| ALE-Bench | 765.62 | 558.1 |
| Aider Polyglot | — | 52.4% |
| LMArena WebDev | 1434 | — |
| SWE-bench Multilingual | 69.7% | — |
| CadEval | — | 42% |
Agentic & Tool Use GPT-4.1 leads
GLM-5: 31.1 (#71), GPT-4.1: 34.7 (#43)
| Benchmark | GLM-5 | GPT-4.1 |
|---|---|---|
| Terminal-Bench | 52.4% | — |
| Berkeley Function Calling Leaderboard | — | 54% |
| τ²-bench Airline | 82.5% | — |
| τ²-bench Banking | 9.8% | — |
| τ²-bench Retail | 73.7% | — |
| τ²-bench Telecom | 86.8% | — |
| Vending-Bench 2 | 4,432 | — |
Reasoning GLM-5 leads
GLM-5: 27.6 (#116), GPT-4.1: 11.7 (#339)
| Benchmark | GLM-5 | GPT-4.1 |
|---|---|---|
| ARC-AGI-2 | 4.9% | 0.4% |
| SimpleBench | 53.2% | 27% |
| Kagi LLM Benchmark | 75% | 52.3% |
| ARC-AGI-1 | 44.7% | 5.5% |
| Chess Puzzles | 10% | 6% |
| LMArena Hard Prompts | 1452 | 1384 |
| Epoch Capabilities Index | 145.83 | 136.78 |
| ForecastBench | 61 | 61.5 |
| NYT Connections (extended) | 74.8% | — |
| EnigmaEval | — | 2.2% |
| DTBench | — | 68.3% |
| LMCA | — | 25.6% |
Math GLM-5 leads
GLM-5: 46.4 (#71), GPT-4.1: 22.3 (#280)
| Benchmark | GLM-5 | GPT-4.1 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 80% | 38.3% |
| LMArena Math | 1440 | 1370 |
| FrontierMath (Feb 2025 set) | 16.4% | 5.5% |
| FrontierMath Tier 4 (v1) | 2.1% | 0% |
| FrontierMath (Tiers 1-3) | — | 6% |
| MathArena Final-Answer Competitions | 65.7% | — |
| Omni-MATH | — | 47.1% |
| MATH Level 5 | — | 83% |
Knowledge GLM-5 leads
GLM-5: 52.3 (#64), GPT-4.1: 37.1 (#160)
| Benchmark | GLM-5 | GPT-4.1 |
|---|---|---|
| GPQA Diamond | 87.8% | 66.9% |
| Vectara Hallucination Rate | 10.1% | 5.6% |
| LMArena Expert | 1454 | 1364 |
| Humanity's Last Exam | — | 5.4% |
| SimpleQA Verified | — | 31.1% |
| MMLU-Pro | — | 81.1% |
| GPQA (HELM) | — | 65.9% |
Multimodal Not comparable
GLM-5: —, GPT-4.1: 38.2 (#67)
| Benchmark | GLM-5 | GPT-4.1 |
|---|---|---|
| LMArena Vision | — | 1211 |
| GeoBench | — | 72% |
Multilingual GLM-5 leads
GLM-5: 53.7 (#58), GPT-4.1: 49.4 (#133)
| Benchmark | GLM-5 | GPT-4.1 |
|---|---|---|
| LMArena Non-English | 1430 | 1370 |
| LMArena Chinese | 1511 | 1382 |
| LMArena French | 1455 | 1382 |
| LMArena German | 1445 | 1381 |
| LMArena Japanese | 1416 | 1319 |
| LMArena Korean | 1423 | 1339 |
| LMArena Russian | 1436 | 1377 |
| LMArena Spanish | 1454 | 1376 |
Instruction Following GLM-5 leads
GLM-5: 75.2 (#67), GPT-4.1: 71.3 (#153)
| Benchmark | GLM-5 | GPT-4.1 |
|---|---|---|
| LMArena Instruction Following | 1428 | 1367 |
| IFEval | — | 83.8% |
Long Context GLM-5 leads
GLM-5: 44.7 (#60), GPT-4.1: 40.0 (#163)
| Benchmark | GLM-5 | GPT-4.1 |
|---|---|---|
| LMArena Longer Query | 1446 | 1385 |
| Fiction.LiveBench | — | 63.9% |
| CL-bench | 18.7% | — |
Writing & Preference GLM-5 leads
GLM-5: 66.0 (#38), GPT-4.1: 57.6 (#125)
| Benchmark | GLM-5 | GPT-4.1 |
|---|---|---|
| LMArena Text | 1446 | 1383 |
| LMArena Creative Writing | 1439 | 1363 |
| EQ-Bench Creative Writing | 1601 | 1420 |
| LMArena Multi-Turn | 1456 | 1398 |
| WildBench | — | 85.4% |
Frequently asked questions
Is GLM-5 better than GPT-4.1?
GLM-5 is the stronger model overall, scoring 46.1 to 35.9 on the Noometry Index.
Which is cheaper, GLM-5 or GPT-4.1?
GLM-5 is cheaper. It lists at $1 per million input tokens and $3.20 per million output tokens; GPT-4.1 lists at $2 and $8.
Is GLM-5 or GPT-4.1 better for coding?
GLM-5 scores higher on coding benchmarks: 49.0 versus 34.4 in the Noometry coding category.
Which has the bigger context window?
GPT-4.1 does, with 1.05M tokens against 205K.
How many benchmarks do GLM-5 and GPT-4.1 share?
34 benchmarks have published results for both models. GLM-5 has 45 scored results on Noometry and GPT-4.1 has 52.