Model comparison
Claude Sonnet 4 vs GLM-5.2
GLM-5.2 is the stronger model overall, scoring 51.1 to 40.8 on the Noometry Index.
Last verified . 31 shared benchmarks.
Summary
- They share 31 benchmarks with published results for both. Claude Sonnet 4 scores higher in 1 category and GLM-5.2 in 8 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where GLM-5.2 leads 42.3 to 22.9.
- The biggest single-benchmark swing is ARC-AGI-1: 40% for Claude Sonnet 4 and 77% for GLM-5.2.
- GLM-5.2 is cheaper at $1.40 / $4.40 per million input/output tokens, against $3 / $15 for Claude Sonnet 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 Sonnet 4 | GLM-5.2 | |
|---|---|---|
| Provider | Anthropic | Z.ai (Zhipu) |
| Noometry Index | 40.8 | 51.1 |
| Released | 2025-05-22 | 2026-06-13 |
| Weights | Proprietary | Open |
| Context window | 200K | 1M |
| Max output | 64K | 131K |
| Input $ / M tokens | $3 | $1.40 |
| Output $ / M tokens | $15 | $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 Sonnet 4: 43.5 (#88), GLM-5.2: 51.3 (#41)
| Benchmark | Claude Sonnet 4 | GLM-5.2 |
|---|---|---|
| SciCode | 40% | 50.5% |
| WeirdML | 46.1% | 70.1% |
| LMArena Coding | 1414 | 1485 |
| ALE-Bench | 655.35 | 1,047 |
| SWE-bench Verified | — | 78.7% |
| DeepSWE | — | 43.8% |
| FrontierCode | — | 24.5% |
| SWE-bench Verified (bash only) | 64.9% | — |
| Aider Polyglot | 61.3% | — |
| LMArena WebDev | — | 1603 |
| GSO | 4.9% | — |
Agentic & Tool Use Claude Sonnet 4 leads
Claude Sonnet 4: 38.5 (#31), GLM-5.2: 32.4 (#63)
| Benchmark | Claude Sonnet 4 | GLM-5.2 |
|---|---|---|
| APEX-Agents | — | 45.2% |
| TheAgentCompany | 33.1% | — |
| τ²-bench Banking | — | 37.1% |
| Cybench | 35% | — |
| DeepResearch Bench | 46.6% | — |
| OSWorld | 43.9% | — |
| PostTrainBench | — | 31.7% |
| GBAEval | — | 0% |
| METR Time Horizons | 62% | — |
| Vending-Bench 2 | — | 8,314 |
Reasoning GLM-5.2 leads
Claude Sonnet 4: 22.9 (#187), GLM-5.2: 42.3 (#52)
| Benchmark | Claude Sonnet 4 | GLM-5.2 |
|---|---|---|
| ARC-AGI-2 | 5.9% | 22.8% |
| SimpleBench | 45.5% | 58.8% |
| Kagi LLM Benchmark | 73% | 62.6% |
| ARC-AGI-1 | 40% | 77% |
| CritPt | 0.3% | 20.9% |
| LMArena Hard Prompts | 1372 | 1480 |
| DTBench | 77.1% | 93.6% |
| LMCA | 29% | 45.8% |
| Epoch Capabilities Index | 141.69 | 151.78 |
| NYT Connections (extended) | — | 74.3% |
| Chess Puzzles | — | 21% |
| EnigmaEval | 3.1% | — |
| EBR-Bench | — | 9.5% |
| Mystery Game Puzzles | — | 19% |
| Surface Evolver Bench | — | 55.6% |
| ForecastBench | 60.2 | — |
Math GLM-5.2 leads
Claude Sonnet 4: 43.3 (#80), GLM-5.2: 55.7 (#43)
| Benchmark | Claude Sonnet 4 | GLM-5.2 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 71.1% | 86.4% |
| LMArena Math | 1375 | 1482 |
| FrontierMath (Tiers 1-3) | — | 59.2% |
| FrontierMath Tier 4 | — | 29.3% |
| MathArena Final-Answer Competitions | — | 67.6% |
| ProofBench | — | 35% |
| Omni-MATH | 60.2% | — |
| MATH Level 5 | 84.4% | — |
| FrontierMath (Feb 2025 set) | 4.1% | — |
| FrontierMath Tier 4 (v1) | 0% | — |
Knowledge GLM-5.2 leads
Claude Sonnet 4: 41.8 (#108), GLM-5.2: 57.1 (#40)
| Benchmark | Claude Sonnet 4 | GLM-5.2 |
|---|---|---|
| GPQA Diamond | 79.2% | 91.9% |
| LMArena Expert | 1372 | 1486 |
| Humanity's Last Exam | 7.8% | — |
| SimpleQA Verified | — | 34.2% |
| MMLU-Pro | 84.3% | — |
| Confabulations | 13.2% | — |
| Vectara Hallucination Rate | 10.3% | — |
| GPQA (HELM) | 70.6% | — |
Multimodal Not comparable
Claude Sonnet 4: 26.2 (#121), GLM-5.2: —
| Benchmark | Claude Sonnet 4 | GLM-5.2 |
|---|---|---|
| LMArena Vision | 1191 | — |
| GeoBench | 37% | — |
| VPCT | 34% | — |
| MindCube | 44.8% | — |
Multilingual GLM-5.2 leads
Claude Sonnet 4: 46.7 (#156), GLM-5.2: 55.8 (#26)
| Benchmark | Claude Sonnet 4 | GLM-5.2 |
|---|---|---|
| LMArena Non-English | 1333 | 1459 |
| LMArena Chinese | 1350 | 1519 |
| LMArena French | 1363 | 1479 |
| LMArena German | 1331 | 1468 |
| LMArena Japanese | 1302 | 1451 |
| LMArena Korean | 1291 | 1445 |
| LMArena Russian | 1355 | 1466 |
| LMArena Spanish | 1357 | 1477 |
Instruction Following GLM-5.2 leads
Claude Sonnet 4: 71.7 (#145), GLM-5.2: 76.9 (#34)
| Benchmark | Claude Sonnet 4 | GLM-5.2 |
|---|---|---|
| LMArena Instruction Following | 1376 | 1465 |
| IFEval | 84% | — |
Long Context GLM-5.2 leads
Claude Sonnet 4: 33.7 (#259), GLM-5.2: 45.3 (#43)
| Benchmark | Claude Sonnet 4 | GLM-5.2 |
|---|---|---|
| LMArena Longer Query | 1398 | 1479 |
| Fiction.LiveBench | 46.9% | — |
Writing & Preference GLM-5.2 leads
Claude Sonnet 4: 57.1 (#132), GLM-5.2: 70.4 (#21)
| Benchmark | Claude Sonnet 4 | GLM-5.2 |
|---|---|---|
| LMArena Text | 1351 | 1470 |
| LMArena Creative Writing | 1345 | 1462 |
| EQ-Bench Creative Writing | 1483 | 1757 |
| LMArena Multi-Turn | 1376 | 1469 |
| Short-Story Creative Writing | 81.4% | — |
| WildBench | 83.8% | — |
| EQ-Bench 4 | — | 1222 |
Frequently asked questions
Is Claude Sonnet 4 better than GLM-5.2?
GLM-5.2 is the stronger model overall, scoring 51.1 to 40.8 on the Noometry Index.
Which is cheaper, Claude Sonnet 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 Sonnet 4 lists at $3 and $15.
Is Claude Sonnet 4 or GLM-5.2 better for coding?
GLM-5.2 scores higher on coding benchmarks: 51.3 versus 43.5 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 Sonnet 4 and GLM-5.2 share?
31 benchmarks have published results for both models. Claude Sonnet 4 has 58 scored results on Noometry and GLM-5.2 has 51.