Model comparison
Claude 3.7 Sonnet vs GLM-5.2
GLM-5.2 is the stronger model overall, scoring 51.1 to 39.5 on the Noometry Index.
Last verified . 25 shared benchmarks.
Summary
- They share 25 benchmarks with published results for both. Claude 3.7 Sonnet scores higher in 2 categories and GLM-5.2 in 7 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where GLM-5.2 leads 42.3 to 18.6.
- The biggest single-benchmark swing is ARC-AGI-1: 28.6% for Claude 3.7 Sonnet and 77% for GLM-5.2.
- GLM-5.2 has downloadable open weights; the other is API-only.
Side by side
| Claude 3.7 Sonnet | GLM-5.2 | |
|---|---|---|
| Provider | Anthropic | Z.ai (Zhipu) |
| Noometry Index | 39.5 | 51.1 |
| Released | 2025-02-24 | 2026-06-13 |
| Weights | Proprietary | Open |
| Context window | — | 1M |
| Max output | — | 131K |
| Input $ / M tokens | — | $1.40 |
| Output $ / M tokens | — | $4.40 |
| Results tracked | 58 | 51 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GLM-5.2 leads
Claude 3.7 Sonnet: 40.6 (#136), GLM-5.2: 51.3 (#41)
| Benchmark | Claude 3.7 Sonnet | GLM-5.2 |
|---|---|---|
| SWE-bench Verified | 61% | 78.7% |
| LMArena Coding | 1361 | 1485 |
| DeepSWE | — | 43.8% |
| FrontierCode | — | 24.5% |
| SWE-bench Verified (bash only) | 52.8% | — |
| Aider Polyglot | 64.9% | — |
| LMArena WebDev | — | 1603 |
| SciCode | — | 50.5% |
| GSO | 3.8% | — |
| WeirdML | — | 70.1% |
| LiveBench Coding | 74.5% | — |
| CadEval | 54% | — |
| ALE-Bench | — | 1,047 |
Agentic & Tool Use Claude 3.7 Sonnet leads
Claude 3.7 Sonnet: 34.1 (#50), GLM-5.2: 32.4 (#63)
| Benchmark | Claude 3.7 Sonnet | GLM-5.2 |
|---|---|---|
| APEX-Agents | — | 45.2% |
| TheAgentCompany | 30.9% | — |
| τ²-bench Banking | — | 37.1% |
| Cybench | 20% | — |
| DeepResearch Bench | 43.6% | — |
| OSWorld | 35.8% | — |
| PostTrainBench | — | 31.7% |
| GBAEval | — | 0% |
| METR Time Horizons | 60% | — |
| Vending-Bench 2 | — | 8,314 |
Reasoning GLM-5.2 leads
Claude 3.7 Sonnet: 18.6 (#277), GLM-5.2: 42.3 (#52)
| Benchmark | Claude 3.7 Sonnet | GLM-5.2 |
|---|---|---|
| ARC-AGI-2 | 0.9% | 22.8% |
| SimpleBench | 46.4% | 58.8% |
| ARC-AGI-1 | 28.6% | 77% |
| LMArena Hard Prompts | 1333 | 1480 |
| Epoch Capabilities Index | 141.16 | 151.78 |
| Kagi LLM Benchmark | — | 62.6% |
| NYT Connections (extended) | — | 74.3% |
| CritPt | — | 20.9% |
| Chess Puzzles | — | 21% |
| EnigmaEval | 4.2% | — |
| EBR-Bench | — | 9.5% |
| LiveBench Reasoning | 87.8% | — |
| Mystery Game Puzzles | — | 19% |
| DTBench | — | 93.6% |
| LiveBench Data Analysis | 74% | — |
| LMCA | — | 45.8% |
| Surface Evolver Bench | — | 55.6% |
| ForecastBench | 61.8 | — |
| LiveBench | 76.1% | — |
Math GLM-5.2 leads
Claude 3.7 Sonnet: 37.5 (#153), GLM-5.2: 55.7 (#43)
| Benchmark | Claude 3.7 Sonnet | GLM-5.2 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 57.8% | 86.4% |
| LMArena Math | 1337 | 1482 |
| FrontierMath (Tiers 1-3) | — | 59.2% |
| FrontierMath Tier 4 | — | 29.3% |
| MathArena Final-Answer Competitions | — | 67.6% |
| ProofBench | — | 35% |
| Omni-MATH | 33% | — |
| LiveBench Math | 79% | — |
| MATH Level 5 | 91.2% | — |
| FrontierMath (Feb 2025 set) | 4.1% | — |
Knowledge GLM-5.2 leads
Claude 3.7 Sonnet: 39.8 (#130), GLM-5.2: 57.1 (#40)
| Benchmark | Claude 3.7 Sonnet | GLM-5.2 |
|---|---|---|
| GPQA Diamond | 79.7% | 91.9% |
| LMArena Expert | 1321 | 1486 |
| Humanity's Last Exam | 8% | — |
| SimpleQA Verified | — | 34.2% |
| MMLU-Pro | 78.4% | — |
| Confabulations | 14.7% | — |
| GPQA (HELM) | 60.8% | — |
Multimodal Not comparable
Claude 3.7 Sonnet: 33.7 (#95), GLM-5.2: —
| Benchmark | Claude 3.7 Sonnet | GLM-5.2 |
|---|---|---|
| LMArena Vision | 1169 | — |
| GeoBench | 68% | — |
| VPCT | 39% | — |
| SpatialViz-Bench | 33.9% | — |
Multilingual GLM-5.2 leads
Claude 3.7 Sonnet: 44.1 (#179), GLM-5.2: 55.8 (#26)
| Benchmark | Claude 3.7 Sonnet | GLM-5.2 |
|---|---|---|
| LMArena Non-English | 1296 | 1459 |
| LMArena Chinese | 1299 | 1519 |
| LMArena French | 1303 | 1479 |
| LMArena German | 1301 | 1468 |
| LMArena Japanese | 1267 | 1451 |
| LMArena Korean | 1249 | 1445 |
| LMArena Russian | 1311 | 1466 |
| LMArena Spanish | 1298 | 1477 |
Instruction Following GLM-5.2 leads
Claude 3.7 Sonnet: 72.9 (#125), GLM-5.2: 76.9 (#34)
| Benchmark | Claude 3.7 Sonnet | GLM-5.2 |
|---|---|---|
| LMArena Instruction Following | 1352 | 1465 |
| LiveBench Instruction Following | 81.3% | — |
| IFEval | 83.4% | — |
Long Context Claude 3.7 Sonnet leads
Claude 3.7 Sonnet: 50.3 (#10), GLM-5.2: 45.3 (#43)
| Benchmark | Claude 3.7 Sonnet | GLM-5.2 |
|---|---|---|
| LMArena Longer Query | 1373 | 1479 |
| Fiction.LiveBench | 83.3% | — |
Writing & Preference GLM-5.2 leads
Claude 3.7 Sonnet: 54.4 (#150), GLM-5.2: 70.4 (#21)
| Benchmark | Claude 3.7 Sonnet | GLM-5.2 |
|---|---|---|
| LMArena Text | 1314 | 1470 |
| LMArena Creative Writing | 1332 | 1462 |
| EQ-Bench Creative Writing | 1412 | 1757 |
| LMArena Multi-Turn | 1339 | 1469 |
| Short-Story Creative Writing | 81.1% | — |
| WildBench | 81.4% | — |
| EQ-Bench 4 | — | 1222 |
| LiveBench Language | 59.9% | — |
Frequently asked questions
Is Claude 3.7 Sonnet better than GLM-5.2?
GLM-5.2 is the stronger model overall, scoring 51.1 to 39.5 on the Noometry Index.
Is Claude 3.7 Sonnet or GLM-5.2 better for coding?
GLM-5.2 scores higher on coding benchmarks: 51.3 versus 40.6 in the Noometry coding category.
How many benchmarks do Claude 3.7 Sonnet and GLM-5.2 share?
25 benchmarks have published results for both models. Claude 3.7 Sonnet has 58 scored results on Noometry and GLM-5.2 has 51.