GLM-5 vs Qwen3 8
GLM-5 
Qwen3 8 Verdict
- Qwen3 8 (max) leads on quality: 66.5 vs 59.6.
- GLM-5 (thinking) is stronger in instruction following.
- Qwen3 8 (max) is stronger in agents & tools, coding, composite, knowledge, long context, maths, reasoning, human preference, multimodal.
- GLM-5 (thinking) is 1.7× cheaper ($1.55 vs $2.67 per 1M blended).
- GLM-5 (thinking) streams 1.6× faster (61 vs 39 tokens per second).
| Metric | GLM-5 (thinking) | Qwen3 8 (max) |
|---|---|---|
| BenchLeader Index | 59.6 | 66.5 |
| Agents & tools score | 57.9 | 68.3 |
| Coding score | 59.1 | 66.5 |
| Composite score | 64.4 | 74.3 |
| Instruction following score | 70.7 | – |
| Knowledge score | 59.1 | 63.5 |
| Long context score | 64.2 | 66.6 |
| Maths score | 63.2 | 67.0 |
| Reasoning score | 53.3 | 76.3 |
| Human preference score | – | 68.2 |
| Multimodal score | – | 67.4 |
| Blended price $/M | $1.55 | $2.67 |
| Output speed | 61 tok/s | 39 tok/s |
| Time to first answer | 52.4 s | 55.1 s |
| Context window | 205k | 1M |
| SciCode | – | 52.9% |
| ProofBench | – | 58.0% |
| Epoch Capabilities Index | – | 156.6 |
| LMArena Text | – | 1481 |
| LMArena Hard Prompts | – | 1503 |
| LMArena Coding | – | 1522 |
| LMArena WebDev | – | 1671 |
| LMArena Vision | – | 1315 |
| LMArena Agent | – | 3.3 |
| LiveBench | – | 78.5% |
| LiveBench Reasoning | – | 88.2% |
| LiveBench Coding | – | 72.9% |
| LiveBench Agentic Coding | – | 64.7% |
| LiveBench Mathematics | – | 91.3% |
| LiveBench Data Analysis | – | 78.4% |
| LiveBench Language | – | 79.7% |
| LiveBench Instruction Following | – | 74.1% |
| AA Intelligence Index | 27.9 | 45.4 |
| IFBench | 72.3% | – |
| AA-LCR | 75.7% | 80.3% |
| MMMU-Pro | – | 82.8% |
| AA-Omniscience | 0.3 | 12.0 |
| Terminal-Bench Hard | 43.2% | – |
| GPQA Diamond (AA) | 82.0% | 92.8% |
| Humanity's Last Exam (AA) | 29.3% | 43.1% |
| SciCode (AA) | – | 53.2% |
| τ²-Bench Telecom (AA) | 98.3% | – |
| AIME (Vals) | 91.7% | – |
| LiveCodeBench | 81.9% | 87.8% |
| MMLU-Pro | 86.0% | 88.6% |
| IOI | – | 68.9% |
| LegalBench | 84.1% | 83.6% |
| CorpFin | 62.9% | 65.8% |
| TaxEval | 70.0% | 75.5% |
| MedQA | 94.3% | – |
| Terminal-Bench 2.1 (Vals) | – | 67.4% |
| SWE-bench (Vals) | 71.4% | 85.6% |
| GPQA Diamond (Vals) | 83.3% | 93.7% |
| Vals Index | – | 51.8 |
| Kagi LLM Benchmark | 75.0% | – |
| τ²-bench | 81.0% | – |
| CritPt | 2.0% | 20.0% |
| GDPval (AA) | – | 58.2% |
| τ²-Bench Banking (AA) | – | 51.3% |
| APEX-Agents (AA) | 14.4% | – |
| CaseLaw v2 | 52.5% | – |
| Code Migration | – | 24.0% |
| Excel Modeling Benchmark | – | 60.1% |
| Finance Agent v2 | – | 50.6% |
| Harvey's Legal Agent Benchmark | – | 10.4% |
| Legal Research Bench | – | 47.6% |
| MedCode | – | 40.7% |
| MedScribe | – | 85.0% |
| MMMU-Pro (Vals) | – | 88.0% |
| MortgageTax | – | 64.0% |
| MysteryMechanism | – | 23.9% |
| ProgramBench | – | 0.0% |
| Public Benefits Bench | – | 67.1% |
| SAGE | – | 51.3% |
| SkillsBench | – | 42.0% |
| Tax Agent Bench | – | 66.0% |
| Terminal-Bench 2.0 (Vals) | 49.4% | – |
| Terminal-Bench 4.0 (Vals) | – | 24.8% |
| Terminal-Bench Science | – | 1.4% |
| Vals Multimodal Index | – | 65.4% |
| Vibe Code Bench 1-100 | – | 12.8% |
| Vibe Code Bench v1.1 | 23.4% | 64.7% |
| LMArena Maths | – | 1498 |
| LMArena Creative Writing | – | 1472 |
| LMArena Instruction Following | – | 1477 |
| LMArena Multi-turn | – | 1495 |
| LMArena Longer Queries | – | 1492 |
Data as of 2026-09-19. Best configuration of each model; every score links to its source on the model pages.
GLM-5 vs Qwen3 8: questions
- Is GLM-5 better than Qwen3 8?
- Qwen3 8 (max) leads on quality: 66.5 vs 59.6. The BenchLeader Index combines every independent quality benchmark; Qwen3 8 (max) is ahead overall as of 2026-09-19, but check the category scores for your use.
- Is GLM-5 better than Qwen3 8 for coding?
- Qwen3 8 scores higher in coding (67 vs 59 on the category index, where 50 is average).
- Is GLM-5 better than Qwen3 8 for agentic tasks?
- Qwen3 8 scores higher in agentic tasks (68 vs 58 on the category index, where 50 is average).
- Which is cheaper, GLM-5 or Qwen3 8?
- GLM-5 is cheaper: $1.55 against $2.67 per million tokens, blended at three input tokens per output token.
- Which is faster, GLM-5 or Qwen3 8?
- GLM-5 streams faster: 61 against 39 output tokens per second.
- Which has the larger context window?
- Qwen3 8 accepts more context: 1M against 205k tokens.