Model comparison
Claude Opus 4.1 vs GLM-5.2
GLM-5.2 is the stronger model overall, scoring 50.9 to 40.8 on the Noometry Index.
Last verified . 32 shared benchmarks.
Summary
- They share 32 benchmarks with published results for both. Claude Opus 4.1 scores higher in 1 category and GLM-5.2 in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where GLM-5.2 leads 55.4 to 22.0.
- The biggest single-benchmark swing is FrontierMath (Tiers 1-3): 12.6% for Claude Opus 4.1 and 59.2% for GLM-5.2.
- GLM-5.2 is cheaper at $1.40 / $4.40 per million input/output tokens, against $15 / $75 for Claude Opus 4.1.
- GLM-5.2 accepts more context: 1M tokens versus 200K.
- GLM-5.2 has downloadable open weights; the other is API-only.
Side by side
| Claude Opus 4.1 | GLM-5.2 | |
|---|---|---|
| Provider | Anthropic | Z.ai (Zhipu) |
| Noometry Index | 40.8 | 50.9 |
| Released | 2025-08-05 | 2026-06-13 |
| Weights | Proprietary | Open |
| Context window | 200K | 1M |
| Max output | 32K | 131K |
| Input $ / M tokens | $15 | $1.40 |
| Output $ / M tokens | $75 | $4.40 |
| Results tracked | 48 | 50 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GLM-5.2 leads
Claude Opus 4.1: 44.3 (#69), GLM-5.2: 50.9 (#37)
| Benchmark | Claude Opus 4.1 | GLM-5.2 |
|---|---|---|
| SWE-bench Verified | 73.3% | 78.7% |
| LMArena WebDev | 1390 | 1604 |
| WeirdML | 45.9% | 70.1% |
| LMArena Coding | 1479 | 1484 |
| ALE-Bench | 674.77 | 1,047 |
| DeepSWE | — | 43.8% |
| FrontierCode | — | 24.5% |
| SciCode | — | 50.5% |
| AlgoTune | 1.34 | — |
Agentic & Tool Use Claude Opus 4.1 leads
Claude Opus 4.1: 34.8 (#38), GLM-5.2: 31.6 (#63)
| Benchmark | Claude Opus 4.1 | GLM-5.2 |
|---|---|---|
| Terminal-Bench | 38% | — |
| GDPval | 43.6% | — |
| τ²-bench Banking | — | 37.1% |
| Cybench | 42% | — |
| DeepResearch Bench | 48.3% | — |
| PostTrainBench | — | 31.7% |
| GBAEval | — | 0% |
| LMArena Search | 1148 | — |
| METR Time Horizons | 66.8% | — |
| Vending-Bench 2 | — | 8,314 |
Reasoning GLM-5.2 leads
Claude Opus 4.1: 31.6 (#70), GLM-5.2: 42.1 (#46)
| Benchmark | Claude Opus 4.1 | GLM-5.2 |
|---|---|---|
| SimpleBench | 60% | 58.8% |
| Chess Puzzles | 7% | 21% |
| EBR-Bench | 7.9% | 9.5% |
| LMArena Hard Prompts | 1443 | 1480 |
| Mystery Game Puzzles | 21% | 19% |
| DTBench | 80% | 93.6% |
| LMCA | 37.1% | 45.8% |
| Epoch Capabilities Index | 144.11 | 151.76 |
| ARC-AGI-2 | — | 22.8% |
| Kagi LLM Benchmark | — | 62.6% |
| NYT Connections (extended) | — | 74.3% |
| ARC-AGI-1 | — | 77% |
| CritPt | — | 20.9% |
| EnigmaEval | 7.2% | — |
| Surface Evolver Bench | — | 55.6% |
| ForecastBench | 62 | — |
Math GLM-5.2 leads
Claude Opus 4.1: 22.0 (#267), GLM-5.2: 55.4 (#35)
| Benchmark | Claude Opus 4.1 | GLM-5.2 |
|---|---|---|
| FrontierMath (Tiers 1-3) | 12.6% | 59.2% |
| FrontierMath Tier 4 | 2.4% | 29.3% |
| OTIS Mock AIME 2024-2025 | 68.9% | 86.4% |
| LMArena Math | 1430 | 1482 |
| MathArena Final-Answer Competitions | — | 67.6% |
| ProofBench | — | 35% |
| FrontierMath (Feb 2025 set) | 7.2% | — |
| FrontierMath Tier 4 (v1) | 4.2% | — |
Knowledge GLM-5.2 leads
Claude Opus 4.1: 41.9 (#93), GLM-5.2: 57.1 (#36)
| Benchmark | Claude Opus 4.1 | GLM-5.2 |
|---|---|---|
| GPQA Diamond | 77.3% | 91.9% |
| LMArena Expert | 1439 | 1484 |
| Humanity's Last Exam | 11.5% | — |
| SimpleQA Verified | — | 34.2% |
| Confabulations | 17.1% | — |
| Vectara Hallucination Rate | 11.8% | — |
Multimodal Not comparable
Claude Opus 4.1: 26.2 (#114), GLM-5.2: —
| Benchmark | Claude Opus 4.1 | GLM-5.2 |
|---|---|---|
| VPCT | 35% | — |
Multilingual GLM-5.2 leads
Claude Opus 4.1: 52.1 (#90), GLM-5.2: 56.0 (#25)
| Benchmark | Claude Opus 4.1 | GLM-5.2 |
|---|---|---|
| LMArena Non-English | 1405 | 1459 |
| LMArena Chinese | 1427 | 1518 |
| LMArena French | 1432 | 1476 |
| LMArena German | 1412 | 1470 |
| LMArena Japanese | 1378 | 1451 |
| LMArena Korean | 1381 | 1445 |
| LMArena Russian | 1421 | 1467 |
| LMArena Spanish | 1448 | 1477 |
Instruction Following GLM-5.2 leads
Claude Opus 4.1: 75.8 (#53), GLM-5.2: 77.1 (#32)
| Benchmark | Claude Opus 4.1 | GLM-5.2 |
|---|---|---|
| LMArena Instruction Following | 1436 | 1465 |
Long Context Too close to call
Claude Opus 4.1: 44.5 (#60), GLM-5.2: 45.3 (#43)
| Benchmark | Claude Opus 4.1 | GLM-5.2 |
|---|---|---|
| LMArena Longer Query | 1455 | 1479 |
Writing & Preference GLM-5.2 leads
Claude Opus 4.1: 62.6 (#70), GLM-5.2: 70.7 (#20)
| Benchmark | Claude Opus 4.1 | GLM-5.2 |
|---|---|---|
| LMArena Text | 1419 | 1470 |
| LMArena Creative Writing | 1412 | 1465 |
| LMArena Multi-Turn | 1444 | 1468 |
| Short-Story Creative Writing | 84.7% | — |
| EQ-Bench Creative Writing | — | 1757 |
| EQ-Bench 4 | — | 1222 |
Frequently asked questions
Is Claude Opus 4.1 better than GLM-5.2?
GLM-5.2 is the stronger model overall, scoring 50.9 to 40.8 on the Noometry Index.
Which is cheaper, Claude Opus 4.1 or GLM-5.2?
GLM-5.2 is cheaper. It lists at $1.40 per million input tokens and $4.40 per million output tokens; Claude Opus 4.1 lists at $15 and $75.
Is Claude Opus 4.1 or GLM-5.2 better for coding?
GLM-5.2 scores higher on coding benchmarks: 50.9 versus 44.3 in the Noometry coding category.
Which has the bigger context window?
GLM-5.2 does, with 1M tokens against 200K.
How many benchmarks do Claude Opus 4.1 and GLM-5.2 share?
32 benchmarks have published results for both models. Claude Opus 4.1 has 48 scored results on Noometry and GLM-5.2 has 50.