GPT-5.6 Terra vs Qwen3.8 Max
GPT-5.6 Terra 
Qwen3.8 Max Verdict
- GPT-5.6 Terra (max) and Qwen3.8 Max (max) are level on quality (63.7 vs 64.2).
- GPT-5.6 Terra (max) is stronger in agents & tools, coding, instruction following, long context, maths.
- Qwen3.8 Max (max) is stronger in composite, knowledge, multimodal, reasoning, human preference.
- Qwen3.8 Max (max) is 1.5× cheaper ($3.00 vs $4.50 per 1M blended).
- GPT-5.6 Terra (max) streams 2.9× faster (108 vs 37 tokens per second).
| Metric | GPT-5.6 Terra (max) | Qwen3.8 Max (max) |
|---|---|---|
| BenchLeader Index | 63.7 | 64.2 |
| Agents & tools score | 58.8 | 58.2 |
| Coding score | 64.7 | 64.6 |
| Composite score | 68.1 | 70.7 |
| Instruction following score | 69.8 | – |
| Knowledge score | 57.5 | 62.7 |
| Long context score | 66.8 | 65.4 |
| Maths score | 69.3 | 65.1 |
| Multimodal score | 63.9 | 66.2 |
| Reasoning score | 70.4 | 72.2 |
| Human preference score | – | 68.0 |
| Blended price $/M | $4.50 | $3.00 |
| Output speed | 108 tok/s | 37 tok/s |
| Time to first answer | 113.7 s | 58.9 s |
| Context window | 1.1M | 1M |
| GPQA Diamond | 93.3% | – |
| FrontierMath Tiers 1–3 | 86.0% | – |
| FrontierMath Tier 4 | 70.7% | – |
| OTIS Mock AIME | 99.7% | – |
| SimpleQA Verified | 43.2% | – |
| Terminal-Bench | 21.5% | 27.0% |
| SciCode | 55.0% | 53.2% |
| APEX-Agents | 58.2% | 63.3% |
| ProofBench | – | 58.0% |
| Epoch Capabilities Index | – | 156.4 |
| LMArena Text | – | 1483 |
| LMArena Hard Prompts | – | 1504 |
| LMArena Coding | – | 1524 |
| LMArena WebDev | – | 1672 |
| LMArena Vision | – | 1314 |
| LMArena Agent | – | 2.3 |
| LiveBench | 77.9% | 78.5% |
| LiveBench Reasoning | 90.6% | 88.2% |
| LiveBench Coding | 78.3% | 72.9% |
| LiveBench Agentic Coding | 55.0% | 64.7% |
| LiveBench Mathematics | 94.9% | 91.3% |
| LiveBench Data Analysis | 79.3% | 78.4% |
| LiveBench Language | 82.9% | 79.7% |
| LiveBench Instruction Following | 64.6% | 74.1% |
| AA Intelligence Index v4.3.2 | 42.1 | 45.4 |
| IFBench | 71.2% | – |
| AA-LCR | 83.0% | 80.3% |
| MMMU-Pro | 80.7% | 82.8% |
| AA-Omniscience | 0.1 | 12.0 |
| Terminal-Bench Hard | 57.6% | – |
| GPQA Diamond (AA) | 92.5% | 92.8% |
| Humanity's Last Exam (AA) | 42.9% | 43.1% |
| SciCode (AA) | 55.0% | 53.2% |
| τ²-Bench Telecom (AA) | 86.3% | – |
| LiveCodeBench | – | 87.8% |
| MMLU-Pro | – | 88.6% |
| IOI | 87.6% | 68.9% |
| LegalBench | 85.1% | 83.6% |
| CorpFin | – | 65.8% |
| TaxEval | – | 75.5% |
| Terminal-Bench 2.1 (Vals) | 77.5% | 67.4% |
| SWE-bench (Vals) | 95.4% | 85.6% |
| GPQA Diamond (Vals) | – | 93.7% |
| Vals Index | 53.1 | 48.3 |
| ARC-AGI-1 | 96.5% | – |
| ARC-AGI-2 | 83.9% | – |
| ARC-AGI-3 | 0.8% | – |
| CritPt | 30.0% | 20.0% |
| GDPval-AA v2.1 | 47.7% | 58.6% |
| τ³-Banking (AA) | 40.2% | 51.3% |
| ITBench SRE (AA) | 51.0% | 40.3% |
| Analyst Agent (AA) | – | 45.0% |
| APEX-Agents (AA) | 38.9% | 42.4% |
| Code Migration | 47.8% | 24.0% |
| CyberBench | 72.1% | 28.6% |
| Excel Modeling Benchmark | 66.2% | 60.1% |
| Finance Agent v2 | 54.4% | 50.6% |
| Harvey's Legal Agent Benchmark | 0.8% | 10.4% |
| Legal Research Bench | 41.4% | 47.6% |
| MedCode | – | 40.7% |
| MedScribe | – | 85.0% |
| MMMU-Pro (Vals) | 86.5% | 88.0% |
| MortgageTax | – | 64.0% |
| MysteryMechanism | – | 23.9% |
| ProgramBench | 0.5% | 0.0% |
| Public Benefits Bench | 62.4% | 67.1% |
| SAGE | 47.0% | 51.3% |
| SkillsBench | 58.9% | 42.0% |
| Tax Agent Bench | 65.2% | 66.0% |
| Terminal-Bench 4.0 (Vals) | 22.7% | 34.3% |
| Terminal-Bench Science | 10.0% | 1.4% |
| Vals Multimodal Index | – | 65.4% |
| Vibe Code Bench 1-100 | 14.8% | 12.8% |
| Vibe Code Bench v1.1 | 74.6% | 64.7% |
| LMArena Maths | – | 1497 |
| LMArena Creative Writing | – | 1470 |
| LMArena Instruction Following | – | 1474 |
| LMArena Multi-turn | – | 1492 |
| LMArena Longer Queries | – | 1492 |
| Chess Puzzles | 54.0% | – |
| Mystery Game Puzzles | 35.0% | – |
| BALROG | 53.2% | – |
| DeepSWE v1.1 | 69.6% | – |
| LMCA | 55.0% | – |
| DTBench | 93.3% | – |
| CursorBench | 41.3% | – |
| ALE-Bench | 1951.4 | – |
| Terminal-Bench 4.0 (AA) | 35.4% | 38.9% |
| Terminal-Bench 2.1 (AA) | 88.0% | 88.8% |
| AutomationBench | – | 56.2% |
| GDP.pdf | – | 22.8% |
| MLCR | – | 20.0% |
| EnterpriseOps-Gym | – | 47.6% |
| AA-Omniscience: accuracy | 46.8% | 31.9% |
| AA-Omniscience: non-hallucination | 12.1% | 71.2% |
| AA-Briefcase v1.1 | – | 1617 |
Data as of 2026-10-11. Best configuration of each model; every score links to its source on the model pages.
GPT-5.6 Terra vs Qwen3.8 Max: questions
- Is GPT-5.6 Terra better than Qwen3.8 Max?
- GPT-5.6 Terra (max) and Qwen3.8 Max (max) are level on quality (63.7 vs 64.2). The BenchLeader Index combines every independent quality benchmark; Qwen3.8 Max (max) is ahead overall as of 2026-10-11, but check the category scores for your use.
- Is GPT-5.6 Terra better than Qwen3.8 Max for coding?
- GPT-5.6 Terra scores higher in coding (65 vs 65 on the category index, where 50 is average).
- Is GPT-5.6 Terra better than Qwen3.8 Max for agentic tasks?
- GPT-5.6 Terra scores higher in agentic tasks (59 vs 58 on the category index, where 50 is average).
- Which is cheaper, GPT-5.6 Terra or Qwen3.8 Max?
- Qwen3.8 Max is cheaper: $3.00 against $4.50 per million tokens, blended at three input tokens per output token.
- Which is faster, GPT-5.6 Terra or Qwen3.8 Max?
- GPT-5.6 Terra streams faster: 108 against 37 output tokens per second.
- Which has the larger context window?
- GPT-5.6 Terra accepts more context: 1.1M against 1M tokens.