Model comparison
Claude Opus 4.7 vs GLM-5.2
Claude Opus 4.7 is the stronger model overall, scoring 58.3 to 51.1 on the Noometry Index. GLM-5.2 costs 4.7× less per token, which makes it the better buy when Claude Opus 4.7's lead doesn't matter for your workload.
Last verified . 49 shared benchmarks.
Summary
- They share 49 benchmarks with published results for both. Claude Opus 4.7 scores higher in 9 categories and GLM-5.2 in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in agentic & tool use, where Claude Opus 4.7 leads 47.9 to 32.4.
- The biggest single-benchmark swing is ARC-AGI-2: 75.8% for Claude Opus 4.7 and 22.8% for GLM-5.2.
- GLM-5.2 is cheaper at $1.40 / $4.40 per million input/output tokens, against $5 / $25 for Claude Opus 4.7.
- GLM-5.2 has downloadable open weights; the other is API-only.
Side by side
| Claude Opus 4.7 | GLM-5.2 | |
|---|---|---|
| Provider | Anthropic | Z.ai (Zhipu) |
| Noometry Index | 58.3 | 51.1 |
| Released | 2026-04-14 | 2026-06-13 |
| Weights | Proprietary | Open |
| Context window | 1M | 1M |
| Max output | 128K | 131K |
| Input $ / M tokens | $5 | $1.40 |
| Output $ / M tokens | $25 | $4.40 |
| Results tracked | 66 | 51 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Claude Opus 4.7 leads
Claude Opus 4.7: 59.6 (#13), GLM-5.2: 51.3 (#41)
| Benchmark | Claude Opus 4.7 | GLM-5.2 |
|---|---|---|
| SWE-bench Verified | 83.5% | 78.7% |
| FrontierCode | 38.5% | 24.5% |
| LMArena WebDev | 1558 | 1603 |
| SciCode | 54.5% | 50.5% |
| WeirdML | 76.4% | 70.1% |
| LMArena Coding | 1518 | 1485 |
| ALE-Bench | 1,323 | 1,047 |
| DeepSWE | — | 43.8% |
| GSO | 44.1% | — |
| MirrorCode | 31.1% | — |
Agentic & Tool Use Claude Opus 4.7 leads
Claude Opus 4.7: 47.9 (#10), GLM-5.2: 32.4 (#63)
| Benchmark | Claude Opus 4.7 | GLM-5.2 |
|---|---|---|
| APEX-Agents | 49.2% | 45.2% |
| τ²-bench Banking | 40.2% | 37.1% |
| PostTrainBench | 28.6% | 31.7% |
| GBAEval | 43.8% | 0% |
| Vending-Bench 2 | 10,937 | 8,314 |
| Terminal-Bench | 80.2% | — |
| OSWorld 2.0 | 18.2% | — |
| ExploitBench | 26.5% | — |
| GDP.pdf | 21% | — |
| LMArena Search | 1233 | — |
Reasoning Claude Opus 4.7 leads
Claude Opus 4.7: 53.8 (#29), GLM-5.2: 42.3 (#52)
| Benchmark | Claude Opus 4.7 | GLM-5.2 |
|---|---|---|
| ARC-AGI-2 | 75.8% | 22.8% |
| SimpleBench | 61.7% | 58.8% |
| Kagi LLM Benchmark | 80.7% | 62.6% |
| NYT Connections (extended) | 39% | 74.3% |
| ARC-AGI-1 | 93.5% | 77% |
| CritPt | 12% | 20.9% |
| Chess Puzzles | 30% | 21% |
| EBR-Bench | 19% | 9.5% |
| LMArena Hard Prompts | 1506 | 1480 |
| Mystery Game Puzzles | 28% | 19% |
| DTBench | 94.7% | 93.6% |
| LMCA | 52.2% | 45.8% |
| Epoch Capabilities Index | 156.25 | 151.78 |
| Thematic Generalization | 72.8% | — |
| Surface Evolver Bench | — | 55.6% |
| ForecastBench | 60.3 | — |
Math Claude Opus 4.7 leads
Claude Opus 4.7: 66.7 (#26), GLM-5.2: 55.7 (#43)
| Benchmark | Claude Opus 4.7 | GLM-5.2 |
|---|---|---|
| FrontierMath (Tiers 1-3) | 70.2% | 59.2% |
| FrontierMath Tier 4 | 31.7% | 29.3% |
| MathArena Final-Answer Competitions | 73.6% | 67.6% |
| OTIS Mock AIME 2024-2025 | 97.8% | 86.4% |
| ProofBench | 54% | 35% |
| LMArena Math | 1499 | 1482 |
| FrontierMath (Feb 2025 set) | 43.8% | — |
| FrontierMath Tier 4 (v1) | 22.9% | — |
Knowledge Claude Opus 4.7 leads
Claude Opus 4.7: 62.6 (#23), GLM-5.2: 57.1 (#40)
| Benchmark | Claude Opus 4.7 | GLM-5.2 |
|---|---|---|
| GPQA Diamond | 90.2% | 91.9% |
| SimpleQA Verified | 51.7% | 34.2% |
| LMArena Expert | 1521 | 1486 |
| Humanity's Last Exam | 36.2% | — |
| Vectara Hallucination Rate | 12% | — |
Multimodal Not comparable
Claude Opus 4.7: 41.2 (#38), GLM-5.2: —
| Benchmark | Claude Opus 4.7 | GLM-5.2 |
|---|---|---|
| LMArena Vision | 1316 | — |
| Blueprint-Bench 2 | 24.5% | — |
| Furniture Assembly | 33.3% | — |
| LMArena Document | 1495 | — |
Multilingual Claude Opus 4.7 leads
Claude Opus 4.7: 57.3 (#10), GLM-5.2: 55.8 (#26)
| Benchmark | Claude Opus 4.7 | GLM-5.2 |
|---|---|---|
| LMArena Non-English | 1480 | 1459 |
| LMArena Chinese | 1531 | 1519 |
| LMArena French | 1503 | 1479 |
| LMArena German | 1495 | 1468 |
| LMArena Japanese | 1472 | 1451 |
| LMArena Korean | 1464 | 1445 |
| LMArena Russian | 1494 | 1466 |
| LMArena Spanish | 1495 | 1477 |
Instruction Following Claude Opus 4.7 leads
Claude Opus 4.7: 78.4 (#10), GLM-5.2: 76.9 (#34)
| Benchmark | Claude Opus 4.7 | GLM-5.2 |
|---|---|---|
| LMArena Instruction Following | 1498 | 1465 |
Long Context Too close to call
Claude Opus 4.7: 46.2 (#25), GLM-5.2: 45.3 (#43)
| Benchmark | Claude Opus 4.7 | GLM-5.2 |
|---|---|---|
| LMArena Longer Query | 1505 | 1479 |
Writing & Preference Claude Opus 4.7 leads
Claude Opus 4.7: 75.1 (#8), GLM-5.2: 70.4 (#21)
| Benchmark | Claude Opus 4.7 | GLM-5.2 |
|---|---|---|
| LMArena Text | 1490 | 1470 |
| LMArena Creative Writing | 1486 | 1462 |
| EQ-Bench Creative Writing | 1914 | 1757 |
| EQ-Bench 4 | 1311 | 1222 |
| LMArena Multi-Turn | 1505 | 1469 |
Frequently asked questions
Is Claude Opus 4.7 better than GLM-5.2?
Claude Opus 4.7 is the stronger model overall, scoring 58.3 to 51.1 on the Noometry Index. GLM-5.2 costs 4.7× less per token, which makes it the better buy when Claude Opus 4.7's lead doesn't matter for your workload.
Which is cheaper, Claude Opus 4.7 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 4.7 lists at $5 and $25.
Is Claude Opus 4.7 or GLM-5.2 better for coding?
Claude Opus 4.7 scores higher on coding benchmarks: 59.6 versus 51.3 in the Noometry coding category.
Which has the bigger context window?
Both accept 1M tokens.
How many benchmarks do Claude Opus 4.7 and GLM-5.2 share?
49 benchmarks have published results for both models. Claude Opus 4.7 has 66 scored results on Noometry and GLM-5.2 has 51.