Model comparison
Claude Opus 5 vs GLM-5
Claude Opus 5 is the stronger model overall, scoring 67.8 to 46.1 on the Noometry Index. GLM-5 costs 6.5× less per token, which makes it the better buy when Claude Opus 5's lead doesn't matter for your workload.
Last verified . 31 shared benchmarks.
Summary
- They share 31 benchmarks with published results for both. Claude Opus 5 scores higher in 9 categories and GLM-5 in 0 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Claude Opus 5 leads 77.2 to 27.6.
- The biggest single-benchmark swing is ARC-AGI-2: 90.4% for Claude Opus 5 and 4.9% for GLM-5.
- GLM-5 is cheaper at $1 / $3.20 per million input/output tokens, against $5 / $25 for Claude Opus 5.
- Claude Opus 5 accepts more context: 1M tokens versus 205K.
- GLM-5 has downloadable open weights; the other is API-only.
Side by side
| Claude Opus 5 | GLM-5 | |
|---|---|---|
| Provider | Anthropic | Z.ai (Zhipu) |
| Noometry Index | 67.8 | 46.1 |
| Released | 2026-07-24 | 2026-02-11 |
| Weights | Proprietary | Open |
| Context window | 1M | 205K |
| Max output | 128K | 131K |
| Input $ / M tokens | $5 | $1 |
| Output $ / M tokens | $25 | $3.20 |
| Results tracked | 57 | 45 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Claude Opus 5 leads
Claude Opus 5: 67.5 (#5), GLM-5: 49.0 (#52)
| Benchmark | Claude Opus 5 | GLM-5 |
|---|---|---|
| LMArena WebDev | 1691 | 1434 |
| WeirdML | 91.8% | 48.2% |
| LMArena Coding | 1534 | 1461 |
| ALE-Bench | 2,165 | 765.62 |
| SWE-bench Verified | — | 72.1% |
| DeepSWE | 73.6% | — |
| FrontierCode | 53.4% | — |
| SWE-bench Verified (bash only) | — | 72.8% |
| CursorBench | 46.6% | — |
| SWE-bench Multilingual | — | 69.7% |
| FrontierSWE | 52% | — |
| SciCode | 56.4% | — |
Agentic & Tool Use Claude Opus 5 leads
Claude Opus 5: 55.6 (#1), GLM-5: 31.1 (#71)
| Benchmark | Claude Opus 5 | GLM-5 |
|---|---|---|
| τ²-bench Banking | 48.7% | 9.8% |
| Vending-Bench 2 | 11,182 | 4,432 |
| Terminal-Bench | — | 52.4% |
| APEX-Agents | 65.8% | — |
| OSWorld 2.0 | 31.4% | — |
| τ²-bench Airline | — | 82.5% |
| τ²-bench Retail | — | 73.7% |
| τ²-bench Telecom | — | 86.8% |
| PostTrainBench | 35% | — |
| BALROG | 63.4% | — |
| GBAEval | 79.6% | — |
| GDP.pdf | 24% | — |
Reasoning Claude Opus 5 leads
Claude Opus 5: 77.2 (#4), GLM-5: 27.6 (#116)
| Benchmark | Claude Opus 5 | GLM-5 |
|---|---|---|
| ARC-AGI-2 | 90.4% | 4.9% |
| SimpleBench | 80.6% | 53.2% |
| NYT Connections (extended) | 94.3% | 74.8% |
| ARC-AGI-1 | 97.5% | 44.7% |
| Chess Puzzles | 42% | 10% |
| LMArena Hard Prompts | 1526 | 1452 |
| Epoch Capabilities Index | 162.78 | 145.83 |
| Kagi LLM Benchmark | — | 75% |
| CritPt | 29.1% | — |
| EBR-Bench | 45.7% | — |
| Mystery Game Puzzles | 59% | — |
| DTBench | 97.9% | — |
| LMCA | 64.5% | — |
| Bench to the Future 3 | 0.12 | — |
| ForecastBench | — | 61 |
Math Claude Opus 5 leads
Claude Opus 5: 86.2 (#8), GLM-5: 46.4 (#71)
| Benchmark | Claude Opus 5 | GLM-5 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 98.9% | 80% |
| LMArena Math | 1531 | 1440 |
| FrontierMath (Tiers 1-3) | 85.6% | — |
| FrontierMath Tier 4 | 73.2% | — |
| MathArena Final-Answer Competitions | — | 65.7% |
| ProofBench | 99% | — |
| FrontierMath (Feb 2025 set) | — | 16.4% |
| FrontierMath Tier 4 (v1) | — | 2.1% |
Knowledge Claude Opus 5 leads
Claude Opus 5: 66.8 (#9), GLM-5: 52.3 (#64)
| Benchmark | Claude Opus 5 | GLM-5 |
|---|---|---|
| GPQA Diamond | 93.9% | 87.8% |
| LMArena Expert | 1557 | 1454 |
| SimpleQA Verified | 59.9% | — |
| Vectara Hallucination Rate | — | 10.1% |
Multimodal Not comparable
Claude Opus 5: 50.8 (#8), GLM-5: —
| Benchmark | Claude Opus 5 | GLM-5 |
|---|---|---|
| LMArena Vision | 1319 | — |
| Blueprint-Bench 2 | 30.4% | — |
| Furniture Assembly | 60.8% | — |
| LMArena Document | 1516 | — |
Multilingual Claude Opus 5 leads
Claude Opus 5: 58.8 (#4), GLM-5: 53.7 (#58)
| Benchmark | Claude Opus 5 | GLM-5 |
|---|---|---|
| LMArena Non-English | 1501 | 1430 |
| LMArena Chinese | 1574 | 1511 |
| LMArena French | 1519 | 1455 |
| LMArena German | 1524 | 1445 |
| LMArena Japanese | 1516 | 1416 |
| LMArena Korean | 1521 | 1423 |
| LMArena Russian | 1507 | 1436 |
| LMArena Spanish | 1519 | 1454 |
Instruction Following Claude Opus 5 leads
Claude Opus 5: 79.2 (#7), GLM-5: 75.2 (#67)
| Benchmark | Claude Opus 5 | GLM-5 |
|---|---|---|
| LMArena Instruction Following | 1517 | 1428 |
Long Context Claude Opus 5 leads
Claude Opus 5: 46.5 (#21), GLM-5: 44.7 (#60)
| Benchmark | Claude Opus 5 | GLM-5 |
|---|---|---|
| LMArena Longer Query | 1515 | 1446 |
| CL-bench | — | 18.7% |
Writing & Preference Claude Opus 5 leads
Claude Opus 5: 79.2 (#1), GLM-5: 66.0 (#38)
| Benchmark | Claude Opus 5 | GLM-5 |
|---|---|---|
| LMArena Text | 1507 | 1446 |
| LMArena Creative Writing | 1491 | 1439 |
| EQ-Bench Creative Writing | 2133 | 1601 |
| LMArena Multi-Turn | 1499 | 1456 |
| EQ-Bench 4 | 1385 | — |
Frequently asked questions
Is Claude Opus 5 better than GLM-5?
Claude Opus 5 is the stronger model overall, scoring 67.8 to 46.1 on the Noometry Index. GLM-5 costs 6.5× less per token, which makes it the better buy when Claude Opus 5's lead doesn't matter for your workload.
Which is cheaper, Claude Opus 5 or GLM-5?
GLM-5 is cheaper. It lists at $1 per million input tokens and $3.20 per million output tokens; Claude Opus 5 lists at $5 and $25.
Is Claude Opus 5 or GLM-5 better for coding?
Claude Opus 5 scores higher on coding benchmarks: 67.5 versus 49.0 in the Noometry coding category.
Which has the bigger context window?
Claude Opus 5 does, with 1M tokens against 205K.
How many benchmarks do Claude Opus 5 and GLM-5 share?
31 benchmarks have published results for both models. Claude Opus 5 has 57 scored results on Noometry and GLM-5 has 45.