Model comparison
Claude Sonnet 4.6 vs GLM-5
Claude Sonnet 4.6 is the stronger model overall, scoring 50.3 to 46.1 on the Noometry Index. GLM-5 costs 3.9× less per token, which makes it the better buy when Claude Sonnet 4.6's lead doesn't matter for your workload.
Last verified . 35 shared benchmarks.
Summary
- They share 35 benchmarks with published results for both. Claude Sonnet 4.6 scores higher in 7 categories and GLM-5 in 2 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Claude Sonnet 4.6 leads 46.1 to 27.6.
- The biggest single-benchmark swing is ARC-AGI-2: 60.4% for Claude Sonnet 4.6 and 4.9% for GLM-5.
- GLM-5 is cheaper at $1 / $3.20 per million input/output tokens, against $3 / $15 for Claude Sonnet 4.6.
- Claude Sonnet 4.6 accepts more context: 1M tokens versus 205K.
- GLM-5 has downloadable open weights; the other is API-only.
Side by side
| Claude Sonnet 4.6 | GLM-5 | |
|---|---|---|
| Provider | Anthropic | Z.ai (Zhipu) |
| Noometry Index | 50.3 | 46.1 |
| Released | 2026-02-17 | 2026-02-11 |
| Weights | Proprietary | Open |
| Context window | 1M | 205K |
| Max output | 128K | 131K |
| Input $ / M tokens | $3 | $1 |
| Output $ / M tokens | $15 | $3.20 |
| Results tracked | 57 | 45 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GLM-5 leads
Claude Sonnet 4.6: 46.3 (#67), GLM-5: 49.0 (#52)
| Benchmark | Claude Sonnet 4.6 | GLM-5 |
|---|---|---|
| SWE-bench Verified | 75.2% | 72.1% |
| LMArena WebDev | 1522 | 1434 |
| WeirdML | 66.1% | 48.2% |
| LMArena Coding | 1504 | 1461 |
| ALE-Bench | 1,327 | 765.62 |
| DeepSWE | 29.9% | — |
| FrontierCode | 24.3% | — |
| SWE-bench Verified (bash only) | — | 72.8% |
| SWE-bench Multilingual | — | 69.7% |
| SciCode | 46.8% | — |
Agentic & Tool Use Claude Sonnet 4.6 leads
Claude Sonnet 4.6: 39.1 (#28), GLM-5: 31.1 (#71)
| Benchmark | Claude Sonnet 4.6 | GLM-5 |
|---|---|---|
| Terminal-Bench | 53.4% | 52.4% |
| Vending-Bench 2 | 7,204 | 4,432 |
| APEX-Agents | 43% | — |
| OSWorld 2.0 | 9.3% | — |
| τ²-bench Airline | — | 82.5% |
| τ²-bench Banking | — | 9.8% |
| τ²-bench Retail | — | 73.7% |
| τ²-bench Telecom | — | 86.8% |
| DeepResearch Bench | 54.9% | — |
| OSWorld | 72.1% | — |
| ExploitBench | 23.6% | — |
| GBAEval | 48.8% | — |
| GDP.pdf | 18% | — |
| LMArena Search | 1221 | — |
Reasoning Claude Sonnet 4.6 leads
Claude Sonnet 4.6: 46.1 (#45), GLM-5: 27.6 (#116)
| Benchmark | Claude Sonnet 4.6 | GLM-5 |
|---|---|---|
| ARC-AGI-2 | 60.4% | 4.9% |
| NYT Connections (extended) | 80.9% | 74.8% |
| ARC-AGI-1 | 86.5% | 44.7% |
| Chess Puzzles | 13% | 10% |
| LMArena Hard Prompts | 1484 | 1452 |
| Epoch Capabilities Index | 152.24 | 145.83 |
| ForecastBench | 62 | 61 |
| SimpleBench | — | 53.2% |
| Kagi LLM Benchmark | — | 75% |
| CritPt | 3.1% | — |
| Thematic Generalization | 76.3% | — |
| Mystery Game Puzzles | 16% | — |
| DTBench | 89.9% | — |
| LMCA | 46.5% | — |
Math Claude Sonnet 4.6 leads
Claude Sonnet 4.6: 52.9 (#49), GLM-5: 46.4 (#71)
| Benchmark | Claude Sonnet 4.6 | GLM-5 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 85.8% | 80% |
| LMArena Math | 1462 | 1440 |
| FrontierMath (Feb 2025 set) | 32.4% | 16.4% |
| FrontierMath Tier 4 (v1) | 8.3% | 2.1% |
| MathArena Final-Answer Competitions | — | 65.7% |
| ProofBench | 45% | — |
Knowledge Too close to call
Claude Sonnet 4.6: 51.7 (#65), GLM-5: 52.3 (#64)
| Benchmark | Claude Sonnet 4.6 | GLM-5 |
|---|---|---|
| GPQA Diamond | 87.4% | 87.8% |
| Vectara Hallucination Rate | 10.6% | 10.1% |
| LMArena Expert | 1500 | 1454 |
| SimpleQA Verified | 35.5% | — |
Multimodal Not comparable
Claude Sonnet 4.6: 38.0 (#68), GLM-5: —
| Benchmark | Claude Sonnet 4.6 | GLM-5 |
|---|---|---|
| LMArena Vision | 1283 | — |
| Blueprint-Bench 2 | 6.7% | — |
| LMArena Document | 1482 | — |
Multilingual Too close to call
Claude Sonnet 4.6: 54.4 (#41), GLM-5: 53.7 (#58)
| Benchmark | Claude Sonnet 4.6 | GLM-5 |
|---|---|---|
| LMArena Non-English | 1440 | 1430 |
| LMArena Chinese | 1491 | 1511 |
| LMArena French | 1465 | 1455 |
| LMArena German | 1428 | 1445 |
| LMArena Japanese | 1420 | 1416 |
| LMArena Korean | 1411 | 1423 |
| LMArena Russian | 1440 | 1436 |
| LMArena Spanish | 1464 | 1454 |
Instruction Following Claude Sonnet 4.6 leads
Claude Sonnet 4.6: 77.4 (#25), GLM-5: 75.2 (#67)
| Benchmark | Claude Sonnet 4.6 | GLM-5 |
|---|---|---|
| LMArena Instruction Following | 1475 | 1428 |
Long Context Too close to call
Claude Sonnet 4.6: 45.3 (#44), GLM-5: 44.7 (#60)
| Benchmark | Claude Sonnet 4.6 | GLM-5 |
|---|---|---|
| LMArena Longer Query | 1479 | 1446 |
| CL-bench | — | 18.7% |
Writing & Preference Claude Sonnet 4.6 leads
Claude Sonnet 4.6: 70.2 (#22), GLM-5: 66.0 (#38)
| Benchmark | Claude Sonnet 4.6 | GLM-5 |
|---|---|---|
| LMArena Text | 1458 | 1446 |
| LMArena Creative Writing | 1435 | 1439 |
| EQ-Bench Creative Writing | 1810 | 1601 |
| LMArena Multi-Turn | 1464 | 1456 |
| EQ-Bench 4 | 1207 | — |
Frequently asked questions
Is Claude Sonnet 4.6 better than GLM-5?
Claude Sonnet 4.6 is the stronger model overall, scoring 50.3 to 46.1 on the Noometry Index. GLM-5 costs 3.9× less per token, which makes it the better buy when Claude Sonnet 4.6's lead doesn't matter for your workload.
Which is cheaper, Claude Sonnet 4.6 or GLM-5?
GLM-5 is cheaper. It lists at $1 per million input tokens and $3.20 per million output tokens; Claude Sonnet 4.6 lists at $3 and $15.
Is Claude Sonnet 4.6 or GLM-5 better for coding?
GLM-5 scores higher on coding benchmarks: 49.0 versus 46.3 in the Noometry coding category.
Which has the bigger context window?
Claude Sonnet 4.6 does, with 1M tokens against 205K.
How many benchmarks do Claude Sonnet 4.6 and GLM-5 share?
35 benchmarks have published results for both models. Claude Sonnet 4.6 has 57 scored results on Noometry and GLM-5 has 45.