Claude Sonnet 5 vs GLM-5.2
Claude Sonnet 5 
GLM-5.2 Verdict
- GLM-5.2 (max) leads on quality: 63.9 vs 62.2.
- Claude Sonnet 5 (high) is stronger in knowledge, maths, multimodal.
- GLM-5.2 (max) is stronger in agents & tools, coding, composite, human preference, long context, reasoning, instruction following.
- GLM-5.2 (max) is 1.9× cheaper ($2.15 vs $4.00 per 1M blended).
| Metric | Claude Sonnet 5 (high) | GLM-5.2 (max) |
|---|---|---|
| BenchLeader Index | 62.2 | 63.9 |
| Agents & tools score | 62.0 | 63.8 |
| Coding score | 61.5 | 63.8 |
| Composite score | 69.6 | 72.1 |
| Human preference score | 65.9 | 67.2 |
| Knowledge score | 62.4 | 55.5 |
| Long context score | 64.7 | 65.6 |
| Maths score | 66.9 | 59.0 |
| Multimodal score | 62.6 | – |
| Reasoning score | 69.6 | 72.9 |
| Instruction following score | – | 71.6 |
| Blended price $/M | $4.00 | $2.15 |
| Output speed | 64 tok/s | 71 tok/s |
| Time to first answer | 2.9 s | 31.6 s |
| Context window | 1M | 1M |
| GPQA Diamond | – | 91.9% |
| FrontierMath Tiers 1–3 | – | 59.2% |
| FrontierMath Tier 4 | – | 29.3% |
| OTIS Mock AIME | – | 86.4% |
| SWE-bench Verified (Epoch) | – | 78.7% |
| SimpleQA Verified | – | 34.2% |
| SciCode | 48.6% | 50.5% |
| WeirdML | 68.8% | 70.1% |
| ProofBench | – | 35.0% |
| LMArena Text | 1461 | 1472 |
| LMArena Hard Prompts | 1489 | 1493 |
| LMArena Coding | 1521 | 1510 |
| LMArena WebDev | 1537 | 1592 |
| LMArena Vision | 1277 | – |
| LMArena Agent | 6 | 4.4 |
| AA Intelligence Index | 32.0 | 34.0 |
| IFBench | – | 73.3% |
| AA-LCR | 76.7% | 78.3% |
| AA-Omniscience | -3.7 | 4.4 |
| Terminal-Bench Hard | – | 50.8% |
| GPQA Diamond (AA) | – | 89.5% |
| Humanity's Last Exam (AA) | 35.7% | 41.1% |
| SciCode (AA) | 54.3% | 51.2% |
| τ²-Bench Telecom (AA) | – | 99.1% |
| Terminal-Bench 2.1 (Vals) | – | 67.8% |
| SWE-bench (Vals) | – | 82.8% |
| CritPt | 15.1% | 20.9% |
| GDPval (AA) | 40.4% | 45.3% |
| τ²-Bench Banking (AA) | – | 34.6% |
| ITBench SRE (AA) | – | 42.7% |
| APEX-Agents (AA) | – | 33.7% |
| Code Migration | – | 37.9% |
| Harvey's Legal Agent Benchmark | – | 7.1% |
| Legal Research Bench | – | 31.3% |
| ProgramBench | – | 0.5% |
| SkillsBench | – | 45.1% |
| SREBench | – | 0.0% |
| Vibe Code Bench v1.1 | – | 64.0% |
| LMArena Maths | 1476 | 1480 |
| LMArena Creative Writing | 1437 | 1451 |
| LMArena Instruction Following | 1465 | 1466 |
| LMArena Multi-turn | 1473 | 1469 |
| LMArena Longer Queries | 1482 | 1483 |
| LMArena Document | 1476 | – |
| Chess Puzzles | – | 21.0% |
| EBR-bench | – | 9.5% |
| PostTrainBench | – | 31.7% |
| DeepSWE | 48.2% | 43.8% |
| LMCA | – | 45.8% |
| DTBench | – | 93.6% |
| CursorBench | 56.9% | 55.0% |
| ALE-Bench | 1463.1 | 1010.2 |
Data as of 2026-09-19. Best configuration of each model; every score links to its source on the model pages.
Claude Sonnet 5 vs GLM-5.2: questions
- Is Claude Sonnet 5 better than GLM-5.2?
- GLM-5.2 (max) leads on quality: 63.9 vs 62.2. The BenchLeader Index combines every independent quality benchmark; GLM-5.2 (max) is ahead overall as of 2026-09-19, but check the category scores for your use.
- Is Claude Sonnet 5 better than GLM-5.2 for coding?
- GLM-5.2 scores higher in coding (64 vs 62 on the category index, where 50 is average).
- Is Claude Sonnet 5 better than GLM-5.2 for agentic tasks?
- GLM-5.2 scores higher in agentic tasks (64 vs 62 on the category index, where 50 is average).
- Which is cheaper, Claude Sonnet 5 or GLM-5.2?
- GLM-5.2 is cheaper: $2.15 against $4.00 per million tokens, blended at three input tokens per output token.
- Which is faster, Claude Sonnet 5 or GLM-5.2?
- GLM-5.2 streams faster: 71 against 64 output tokens per second.
- Which has the larger context window?
- Both accept 1M tokens of context.