GPT-5.5 vs Grok 4.3
GPT-5.5 
Grok 4.3 Verdict
- GPT-5.5 leads on quality: 66.7 vs 60.7.
- GPT-5.5 is stronger in agents & tools, coding, composite, human preference, knowledge, long context, multimodal, reasoning.
- Grok 4.3 (medium) is stronger in instruction following.
- Grok 4.3 (medium) is 7.2× cheaper ($1.56 vs $11.25 per 1M blended).
- Grok 4.3 (medium) streams 1.3× faster (112 vs 85 tokens per second).
| Metric | GPT-5.5 | Grok 4.3 (medium) |
|---|---|---|
| BenchLeader Index | 66.7 | 60.7 |
| Agents & tools score | 68.5 | 60.1 |
| Coding score | 57.2 | – |
| Composite score | 77.8 | 60.3 |
| Human preference score | 67.9 | – |
| Instruction following score | 73.6 | 80.0 |
| Knowledge score | 73.9 | 72.1 |
| Long context score | 68.8 | 64.0 |
| Multimodal score | 64.9 | 60.2 |
| Reasoning score | 70.1 | – |
| Blended price $/M | $11.25 | $1.56 |
| Output speed | 85 tok/s | 112 tok/s |
| Time to first answer | 58.6 s | 12.1 s |
| Context window | 1.1M | 1M |
| Terminal-Bench | 84.7% | – |
| SimpleBench | 69.0% | – |
| Remote Labor Index | 6.3% | – |
| APEX-Agents | 38.5% | – |
| FrontierCode | 43.0% | – |
| Epoch Capabilities Index | 159.1 | – |
| LMArena Text | 1477 | – |
| LMArena Hard Prompts | 1498 | – |
| LMArena Coding | 1509 | – |
| LMArena WebDev | 1458 | – |
| LMArena Vision | 1296 | – |
| LMArena Agent | 3 | – |
| AA Intelligence Index | 38.6 | 24.8 |
| IFBench | 75.8% | 83.3% |
| AA-LCR | 84.3% | 75.0% |
| MMMU-Pro | 79.9% | 75.8% |
| AA-Omniscience | 20.5 | 16.7 |
| Terminal-Bench Hard | 60.6% | 30.3% |
| GPQA Diamond (AA) | 93.5% | 89.0% |
| Humanity's Last Exam (AA) | 45.8% | 30.0% |
| SciCode (AA) | 55.8% | – |
| τ²-Bench Telecom (AA) | 93.9% | 91.2% |
| HiL-Bench | 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-5.5 vs Grok 4.3: questions
- Is GPT-5.5 better than Grok 4.3?
- GPT-5.5 leads on quality: 66.7 vs 60.7. 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-5.5 better than Grok 4.3 for agentic tasks?
- GPT-5.5 scores higher in agentic tasks (69 vs 60 on the category index, where 50 is average).
- Which is cheaper, GPT-5.5 or Grok 4.3?
- Grok 4.3 is cheaper: $1.56 against $11.25 per million tokens, blended at three input tokens per output token.
- Which is faster, GPT-5.5 or Grok 4.3?
- Grok 4.3 streams faster: 112 against 85 output tokens per second.
- Which has the larger context window?
- GPT-5.5 accepts more context: 1.1M against 1M tokens.