Model comparison
GLM-5.1 vs GPT-4.1
GLM-5.1 is the stronger model overall, scoring 47.6 to 35.7 on the Noometry Index.
Last verified . 30 shared benchmarks.
Summary
- They share 30 benchmarks with published results for both. GLM-5.1 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-5.1 leads 49.0 to 21.4.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 93.3% for GLM-5.1 and 38.3% for GPT-4.1.
- GLM-5.1 is cheaper at $1.40 / $4.40 per million input/output tokens, against $2 / $8 for GPT-4.1.
- GPT-4.1 accepts more context: 1.05M tokens versus 200K.
- GLM-5.1 has downloadable open weights; the other is API-only.
Side by side
| GLM-5.1 | GPT-4.1 | |
|---|---|---|
| Provider | Z.ai (Zhipu) | OpenAI |
| Noometry Index | 47.6 | 35.7 |
| Released | 2026-04-07 | 2025-04-14 |
| Weights | Open | Proprietary |
| Context window | 200K | 1.05M |
| Max output | 131K | 33K |
| Input $ / M tokens | $1.40 | $2 |
| Output $ / M tokens | $4.40 | $8 |
| Results tracked | 41 | 52 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GLM-5.1 leads
GLM-5.1: 48.5 (#52), GPT-4.1: 34.2 (#228)
| Benchmark | GLM-5.1 | GPT-4.1 |
|---|---|---|
| SWE-bench Verified | 74.2% | 48.5% |
| WeirdML | 57.1% | 39% |
| LMArena Coding | 1486 | 1391 |
| ALE-Bench | 887.1 | 558.1 |
| SWE-bench Verified (bash only) | — | 39.6% |
| Aider Polyglot | — | 52.4% |
| LMArena WebDev | 1509 | — |
| SciCode | 43.8% | — |
| CadEval | — | 42% |
Agentic & Tool Use GPT-4.1 leads
GLM-5.1: 24.8 (#106), GPT-4.1: 34.4 (#40)
| Benchmark | GLM-5.1 | GPT-4.1 |
|---|---|---|
| APEX-Agents | 40.9% | — |
| Berkeley Function Calling Leaderboard | — | 54% |
| ExploitBench | 18.1% | — |
| GBAEval | 0% | — |
| Vending-Bench 2 | 5,634 | — |
Reasoning GLM-5.1 leads
GLM-5.1: 38.2 (#57), GPT-4.1: 11.7 (#334)
| Benchmark | GLM-5.1 | GPT-4.1 |
|---|---|---|
| SimpleBench | 55.1% | 27% |
| Chess Puzzles | 19% | 6% |
| LMArena Hard Prompts | 1472 | 1384 |
| Epoch Capabilities Index | 149.86 | 136.79 |
| ARC-AGI-2 | — | 0.4% |
| Kagi LLM Benchmark | — | 52.3% |
| NYT Connections (extended) | 77.7% | — |
| ARC-AGI-1 | — | 5.5% |
| CritPt | 4.6% | — |
| EnigmaEval | — | 2.2% |
| Thematic Generalization | 69.8% | — |
| DTBench | — | 68.3% |
| LMCA | — | 25.6% |
| ForecastBench | — | 61.5 |
Math GLM-5.1 leads
GLM-5.1: 49.0 (#48), GPT-4.1: 21.4 (#269)
| Benchmark | GLM-5.1 | GPT-4.1 |
|---|---|---|
| FrontierMath (Tiers 1-3) | 36.8% | 6% |
| OTIS Mock AIME 2024-2025 | 93.3% | 38.3% |
| LMArena Math | 1472 | 1369 |
| FrontierMath (Feb 2025 set) | 33.4% | 5.5% |
| FrontierMath Tier 4 (v1) | 12.5% | 0% |
| MathArena Final-Answer Competitions | 67.1% | — |
| ProofBench | 22.2% | — |
| Omni-MATH | — | 47.1% |
| MATH Level 5 | — | 83% |
Knowledge GLM-5.1 leads
GLM-5.1: 55.0 (#45), GPT-4.1: 37.0 (#152)
| Benchmark | GLM-5.1 | GPT-4.1 |
|---|---|---|
| GPQA Diamond | 89.9% | 66.9% |
| SimpleQA Verified | 34% | 31.1% |
| LMArena Expert | 1477 | 1364 |
| Humanity's Last Exam | — | 5.4% |
| MMLU-Pro | — | 81.1% |
| Vectara Hallucination Rate | — | 5.6% |
| GPQA (HELM) | — | 65.9% |
Multimodal Not comparable
GLM-5.1: —, GPT-4.1: 37.5 (#66)
| Benchmark | GLM-5.1 | GPT-4.1 |
|---|---|---|
| LMArena Vision | — | 1210 |
| GeoBench | — | 72% |
Multilingual GLM-5.1 leads
GLM-5.1: 55.2 (#34), GPT-4.1: 49.6 (#128)
| Benchmark | GLM-5.1 | GPT-4.1 |
|---|---|---|
| LMArena Non-English | 1448 | 1370 |
| LMArena Chinese | 1516 | 1382 |
| LMArena French | 1478 | 1383 |
| LMArena German | 1466 | 1380 |
| LMArena Japanese | 1436 | 1320 |
| LMArena Korean | 1423 | 1339 |
| LMArena Russian | 1453 | 1377 |
| LMArena Spanish | 1467 | 1375 |
Instruction Following GLM-5.1 leads
GLM-5.1: 76.5 (#41), GPT-4.1: 71.5 (#148)
| Benchmark | GLM-5.1 | GPT-4.1 |
|---|---|---|
| LMArena Instruction Following | 1451 | 1367 |
| IFEval | — | 83.8% |
Long Context GLM-5.1 leads
GLM-5.1: 44.9 (#51), GPT-4.1: 40.0 (#158)
| Benchmark | GLM-5.1 | GPT-4.1 |
|---|---|---|
| LMArena Longer Query | 1467 | 1385 |
| Fiction.LiveBench | — | 63.9% |
Writing & Preference GLM-5.1 leads
GLM-5.1: 67.2 (#29), GPT-4.1: 57.8 (#120)
| Benchmark | GLM-5.1 | GPT-4.1 |
|---|---|---|
| LMArena Text | 1462 | 1383 |
| LMArena Creative Writing | 1454 | 1363 |
| EQ-Bench Creative Writing | 1592 | 1420 |
| LMArena Multi-Turn | 1473 | 1398 |
| WildBench | — | 85.4% |
Frequently asked questions
Is GLM-5.1 better than GPT-4.1?
GLM-5.1 is the stronger model overall, scoring 47.6 to 35.7 on the Noometry Index.
Which is cheaper, GLM-5.1 or GPT-4.1?
GLM-5.1 is cheaper. It lists at $1.40 per million input tokens and $4.40 per million output tokens; GPT-4.1 lists at $2 and $8.
Is GLM-5.1 or GPT-4.1 better for coding?
GLM-5.1 scores higher on coding benchmarks: 48.5 versus 34.2 in the Noometry coding category.
Which has the bigger context window?
GPT-4.1 does, with 1.05M tokens against 200K.
How many benchmarks do GLM-5.1 and GPT-4.1 share?
30 benchmarks have published results for both models. GLM-5.1 has 41 scored results on Noometry and GPT-4.1 has 52.