Model comparison
Claude Opus 4 vs GLM-5.2
GLM-5.2 is the stronger model overall, scoring 51.1 to 43.1 on the Noometry Index.
Last verified . 30 shared benchmarks.
Summary
- They share 30 benchmarks with published results for both. Claude Opus 4 scores higher in 2 categories and GLM-5.2 in 7 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where GLM-5.2 leads 42.3 to 27.3.
- The biggest single-benchmark swing is ARC-AGI-1: 35.7% for Claude Opus 4 and 77% 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.
- 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 | GLM-5.2 | |
|---|---|---|
| Provider | Anthropic | Z.ai (Zhipu) |
| Noometry Index | 43.1 | 51.1 |
| Released | 2025-05-22 | 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 | 56 | 51 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GLM-5.2 leads
Claude Opus 4: 47.2 (#62), GLM-5.2: 51.3 (#41)
| Benchmark | Claude Opus 4 | GLM-5.2 |
|---|---|---|
| SWE-bench Verified | 70.7% | 78.7% |
| WeirdML | 43.7% | 70.1% |
| LMArena Coding | 1442 | 1485 |
| DeepSWE | — | 43.8% |
| FrontierCode | — | 24.5% |
| SWE-bench Verified (bash only) | 67.6% | — |
| Aider Polyglot | 72% | — |
| LMArena WebDev | — | 1603 |
| SciCode | — | 50.5% |
| GSO | 6.9% | — |
| ALE-Bench | — | 1,047 |
| AlgoTune | 1.33 | — |
Agentic & Tool Use Claude Opus 4 leads
Claude Opus 4: 34.8 (#42), GLM-5.2: 32.4 (#63)
| Benchmark | Claude Opus 4 | GLM-5.2 |
|---|---|---|
| APEX-Agents | — | 45.2% |
| τ²-bench Banking | — | 37.1% |
| Cybench | 38% | — |
| DeepResearch Bench | 46.8% | — |
| PostTrainBench | — | 31.7% |
| GBAEval | — | 0% |
| LMArena Search | 1127 | — |
| METR Time Horizons | 63.9% | — |
| Vending-Bench 2 | — | 8,314 |
Reasoning GLM-5.2 leads
Claude Opus 4: 27.3 (#121), GLM-5.2: 42.3 (#52)
| Benchmark | Claude Opus 4 | GLM-5.2 |
|---|---|---|
| ARC-AGI-2 | 8.6% | 22.8% |
| SimpleBench | 58.8% | 58.8% |
| Kagi LLM Benchmark | 74.3% | 62.6% |
| ARC-AGI-1 | 35.7% | 77% |
| CritPt | 0.3% | 20.9% |
| LMArena Hard Prompts | 1399 | 1480 |
| DTBench | 81.6% | 93.6% |
| LMCA | 37.4% | 45.8% |
| Epoch Capabilities Index | 142.67 | 151.78 |
| NYT Connections (extended) | — | 74.3% |
| Chess Puzzles | — | 21% |
| EnigmaEval | 5.6% | — |
| EBR-Bench | — | 9.5% |
| Mystery Game Puzzles | — | 19% |
| Surface Evolver Bench | — | 55.6% |
| ForecastBench | 61.1 | — |
Math GLM-5.2 leads
Claude Opus 4: 42.0 (#86), GLM-5.2: 55.7 (#43)
| Benchmark | Claude Opus 4 | GLM-5.2 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 64.4% | 86.4% |
| LMArena Math | 1390 | 1482 |
| FrontierMath (Tiers 1-3) | — | 59.2% |
| FrontierMath Tier 4 | — | 29.3% |
| MathArena Final-Answer Competitions | — | 67.6% |
| ProofBench | — | 35% |
| Omni-MATH | 61.6% | — |
| MATH Level 5 | 85% | — |
| FrontierMath (Feb 2025 set) | 4.5% | — |
| FrontierMath Tier 4 (v1) | 4.2% | — |
Knowledge GLM-5.2 leads
Claude Opus 4: 44.0 (#88), GLM-5.2: 57.1 (#40)
| Benchmark | Claude Opus 4 | GLM-5.2 |
|---|---|---|
| GPQA Diamond | 76.3% | 91.9% |
| LMArena Expert | 1386 | 1486 |
| Humanity's Last Exam | 10.7% | — |
| SimpleQA Verified | — | 34.2% |
| MMLU-Pro | 87.5% | — |
| Confabulations | 15.9% | — |
| Vectara Hallucination Rate | 12% | — |
| GPQA (HELM) | 70.8% | — |
Multimodal Not comparable
Claude Opus 4: 31.5 (#106), GLM-5.2: —
| Benchmark | Claude Opus 4 | GLM-5.2 |
|---|---|---|
| LMArena Vision | 1192 | — |
| GeoBench | 49% | — |
| VPCT | 38% | — |
Multilingual GLM-5.2 leads
Claude Opus 4: 48.8 (#138), GLM-5.2: 55.8 (#26)
| Benchmark | Claude Opus 4 | GLM-5.2 |
|---|---|---|
| LMArena Non-English | 1362 | 1459 |
| LMArena Chinese | 1386 | 1519 |
| LMArena French | 1372 | 1479 |
| LMArena German | 1391 | 1468 |
| LMArena Japanese | 1331 | 1451 |
| LMArena Korean | 1321 | 1445 |
| LMArena Russian | 1392 | 1466 |
| LMArena Spanish | 1389 | 1477 |
Instruction Following Too close to call
Claude Opus 4: 77.1 (#28), GLM-5.2: 76.9 (#34)
| Benchmark | Claude Opus 4 | GLM-5.2 |
|---|---|---|
| LMArena Instruction Following | 1406 | 1465 |
| IFEval | 91.8% | — |
Long Context GLM-5.2 leads
Claude Opus 4: 39.6 (#172), GLM-5.2: 45.3 (#43)
| Benchmark | Claude Opus 4 | GLM-5.2 |
|---|---|---|
| LMArena Longer Query | 1422 | 1479 |
| Fiction.LiveBench | 61.1% | — |
Writing & Preference GLM-5.2 leads
Claude Opus 4: 61.2 (#89), GLM-5.2: 70.4 (#21)
| Benchmark | Claude Opus 4 | GLM-5.2 |
|---|---|---|
| LMArena Text | 1377 | 1470 |
| LMArena Creative Writing | 1387 | 1462 |
| EQ-Bench Creative Writing | 1580 | 1757 |
| LMArena Multi-Turn | 1396 | 1469 |
| Short-Story Creative Writing | 83.6% | — |
| WildBench | 85.2% | — |
| EQ-Bench 4 | — | 1222 |
Frequently asked questions
Is Claude Opus 4 better than GLM-5.2?
GLM-5.2 is the stronger model overall, scoring 51.1 to 43.1 on the Noometry Index.
Which is cheaper, Claude Opus 4 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 lists at $15 and $75.
Is Claude Opus 4 or GLM-5.2 better for coding?
GLM-5.2 scores higher on coding benchmarks: 51.3 versus 47.2 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 and GLM-5.2 share?
30 benchmarks have published results for both models. Claude Opus 4 has 56 scored results on Noometry and GLM-5.2 has 51.