GPT-5.3-Codex vs GPT-5.5
GPT-5.3-Codex Verdict
- GPT-5.5 leads on quality: 66.7 vs 62.5.
- GPT-5.5 is stronger in agents & tools, coding, composite, instruction following, knowledge, long context, multimodal, human preference, reasoning.
- GPT-5.3-Codex is 2.3× cheaper ($4.81 vs $11.25 per 1M blended).
- GPT-5.3-Codex streams 1.4× faster (121 vs 85 tokens per second).
| Metric | GPT-5.3-Codex | GPT-5.5 |
|---|---|---|
| BenchLeader Index | 62.5 | 66.7 |
| Agents & tools score | 62.7 | 68.5 |
| Coding score | 54.8 | 57.2 |
| Composite score | 70.1 | 77.8 |
| Instruction following score | 73.2 | 73.6 |
| Knowledge score | 69.3 | 73.9 |
| Long context score | 68.3 | 68.8 |
| Multimodal score | 63.0 | 64.9 |
| Human preference score | – | 67.9 |
| Reasoning score | – | 70.1 |
| Blended price $/M | $4.81 | $11.25 |
| Output speed | 121 tok/s | 85 tok/s |
| Time to first answer | 58.2 s | 58.6 s |
| Context window | 400k | 1.1M |
| Terminal-Bench | 78.4% | 84.7% |
| SimpleBench | – | 69.0% |
| Remote Labor Index | – | 6.3% |
| WeirdML | 79.3% | – |
| APEX-Agents | 31.8% | 38.5% |
| FrontierCode | – | 43.0% |
| Epoch Capabilities Index | 156.6 | 159.1 |
| LMArena Text | – | 1477 |
| LMArena Hard Prompts | – | 1498 |
| LMArena Coding | – | 1509 |
| LMArena WebDev | 1409 | 1458 |
| LMArena Vision | – | 1296 |
| LMArena Agent | – | 3 |
| AA Intelligence Index | 32.5 | 38.6 |
| IFBench | 75.4% | 75.8% |
| AA-LCR | 83.3% | 84.3% |
| MMMU-Pro | 78.5% | 79.9% |
| AA-Omniscience | 10.9 | 20.5 |
| Terminal-Bench Hard | 53.0% | 60.6% |
| GPQA Diamond (AA) | 91.5% | 93.5% |
| Humanity's Last Exam (AA) | 42.5% | 45.8% |
| SciCode (AA) | – | 55.8% |
| τ²-Bench Telecom (AA) | 86.0% | 93.9% |
| HiL-Bench | 4.3% | 39.7% |
| EQ-Bench 4 | – | 1315 |
| Kagi LLM Benchmark | – | 88.8% |
Data as of 2026-09-10. Best configuration of each model; every score links to its source on the model pages.
GPT-5.3-Codex vs GPT-5.5: questions
- Is GPT-5.3-Codex better than GPT-5.5?
- GPT-5.5 leads on quality: 66.7 vs 62.5. The BenchLeader Index combines every independent quality benchmark; GPT-5.5 is ahead overall as of 2026-09-10, but check the category scores for your use.
- Is GPT-5.3-Codex better than GPT-5.5 for coding?
- GPT-5.5 scores higher in coding (57 vs 55 on the category index, where 50 is average).
- Is GPT-5.3-Codex better than GPT-5.5 for agentic tasks?
- GPT-5.5 scores higher in agentic tasks (69 vs 63 on the category index, where 50 is average).
- Which is cheaper, GPT-5.3-Codex or GPT-5.5?
- GPT-5.3-Codex is cheaper: $4.81 against $11.25 per million tokens, blended at three input tokens per output token.
- Which is faster, GPT-5.3-Codex or GPT-5.5?
- GPT-5.3-Codex streams faster: 121 against 85 output tokens per second.
- Which has the larger context window?
- GPT-5.5 accepts more context: 1.1M against 400k tokens.