GPT-5.2-Codex vs GPT-5.6 Terra
GPT-5.2-Codex Verdict
- GPT-5.6 Terra (xhigh) leads on quality: 64.4 vs 60.8.
- GPT-5.2-Codex is stronger in instruction following, knowledge, long context.
- GPT-5.6 Terra (xhigh) is stronger in agents & tools, coding, composite, multimodal, human preference, maths, reasoning.
- They cost about the same ($4.50 per 1M blended).
- GPT-5.6 Terra (xhigh) streams 3.2× faster (79 vs 25 tokens per second).
| Metric | GPT-5.2-Codex | GPT-5.6 Terra (xhigh) |
|---|---|---|
| BenchLeader Index | 60.8 | 64.4 |
| Agents & tools score | 63.5 | 69.8 |
| Coding score | 55.7 | 63.7 |
| Composite score | 56.8 | 77.3 |
| Instruction following score | 75.1 | 65.3 |
| Knowledge score | 63.1 | 61.0 |
| Long context score | 67.8 | 66.0 |
| Multimodal score | 60.7 | 63.0 |
| Human preference score | – | 66.7 |
| Maths score | – | 71.0 |
| Reasoning score | – | 58.6 |
| Blended price $/M | $4.81 | $4.50 |
| Output speed | 25 tok/s | 79 tok/s |
| Time to first answer | 6.7 s | 33.6 s |
| Context window | 400k | 1M |
| Terminal-Bench | 66.5% | – |
| SimpleBench | – | 48.9% |
| SciCode | – | 51.6% |
| APEX-Agents | 27.6% | – |
| ProofBench | – | 74.0% |
| LMArena Text | – | 1466 |
| LMArena Hard Prompts | – | 1491 |
| LMArena Coding | – | 1519 |
| LMArena WebDev | 1339 | 1521 |
| LMArena Vision | – | 1269 |
| LMArena Agent | – | 1.5 |
| LiveBench | 74.0% | – |
| LiveBench Reasoning | 77.7% | – |
| LiveBench Coding | 83.6% | – |
| LiveBench Agentic Coding | 49.4% | – |
| LiveBench Mathematics | 88.8% | – |
| LiveBench Data Analysis | 78.2% | – |
| LiveBench Language | 73.7% | – |
| AA Intelligence Index | 28.5 | 38.2 |
| IFBench | 77.6% | 66.3% |
| AA-LCR | 82.3% | 79.0% |
| MMMU-Pro | 76.3% | 79.5% |
| AA-Omniscience | -2.2 | -3.0 |
| Terminal-Bench Hard | 37.1% | 62.9% |
| GPQA Diamond (AA) | 89.9% | 90.8% |
| Humanity's Last Exam (AA) | 35.7% | 41.9% |
| SciCode (AA) | – | 52.3% |
| τ²-Bench Telecom (AA) | 92.1% | 80.4% |
| LiveCodeBench | 88.0% | 85.9% |
| MMLU-Pro | – | 86.7% |
| IOI | – | 65.3% |
| CorpFin | – | 65.3% |
| TaxEval | – | 76.2% |
| GPQA Diamond (Vals) | – | 90.9% |
| SWE-Bench Pro | 41.0% | – |
| SWE-bench Verified (bash only) | 72.8% | – |
| SWE-bench Verified (any scaffold) | 72.8% | – |
| ARC-AGI-1 | – | 94.0% |
| ARC-AGI-2 | – | 74.2% |
| ARC-AGI-3 | – | 0.7% |
Data as of 2026-09-10. Best configuration of each model; every score links to its source on the model pages.
GPT-5.2-Codex vs GPT-5.6 Terra: questions
- Is GPT-5.2-Codex better than GPT-5.6 Terra?
- GPT-5.6 Terra (xhigh) leads on quality: 64.4 vs 60.8. The BenchLeader Index combines every independent quality benchmark; GPT-5.6 Terra (xhigh) is ahead overall as of 2026-09-10, but check the category scores for your use.
- Is GPT-5.2-Codex better than GPT-5.6 Terra for coding?
- GPT-5.6 Terra scores higher in coding (64 vs 56 on the category index, where 50 is average).
- Is GPT-5.2-Codex better than GPT-5.6 Terra for agentic tasks?
- GPT-5.6 Terra scores higher in agentic tasks (70 vs 64 on the category index, where 50 is average).
- Which is cheaper, GPT-5.2-Codex or GPT-5.6 Terra?
- GPT-5.6 Terra is cheaper: $4.50 against $4.81 per million tokens, blended at three input tokens per output token.
- Which is faster, GPT-5.2-Codex or GPT-5.6 Terra?
- GPT-5.6 Terra streams faster: 79 against 25 output tokens per second.
- Which has the larger context window?
- GPT-5.6 Terra accepts more context: 1M against 400k tokens.