GPT-5.4 vs Grok 4.7
GPT-5.4 
Grok 4.7 Verdict
- GPT-5.4 (xhigh) and Grok 4.7 (high) are level on quality (65.4 vs 65.8).
- GPT-5.4 (xhigh) is stronger in agents & tools, coding, instruction following, long context, maths, multimodal.
- Grok 4.7 (high) is stronger in composite, knowledge, reasoning.
- Grok 4.7 (high) is 1.9× cheaper ($3.00 vs $5.63 per 1M blended).
- GPT-5.4 (xhigh) streams 2.7× faster (148 vs 55 tokens per second).
| Metric | GPT-5.4 (xhigh) | Grok 4.7 (high) |
|---|---|---|
| BenchLeader Index | 65.4 | 65.8 |
| Agents & tools score | 63.4 | – |
| Coding score | 66.6 | – |
| Composite score | 69.3 | 87.5 |
| Instruction following score | 72.1 | – |
| Knowledge score | 60.5 | 72.6 |
| Long context score | 67.4 | 64.8 |
| Maths score | 66.4 | – |
| Multimodal score | 62.9 | – |
| Reasoning score | 72.4 | 76.4 |
| Blended price $/M | $5.63 | $3.00 |
| Output speed | 148 tok/s | 55 tok/s |
| Time to first answer | 150.6 s | 1.4 s |
| Context window | 1.1M | 500k |
| GPQA Diamond | 93.3% | – |
| FrontierMath Tiers 1–3 | 78.6% | – |
| FrontierMath Tier 4 | 49.0% | – |
| OTIS Mock AIME | 95.3% | – |
| SimpleQA Verified | 45.1% | – |
| Humanity's Last Exam | 36.2% | – |
| SciCode | 56.6% | – |
| WeirdML | 77.7% | – |
| APEX-Agents | 36.0% | – |
| ProofBench | 56.0% | – |
| GSO-Bench | 31.4% | – |
| LiveBench | 78.0% | – |
| LiveBench Reasoning | 88.1% | – |
| LiveBench Coding | 77.5% | – |
| LiveBench Agentic Coding | 53.8% | – |
| LiveBench Mathematics | 94.2% | – |
| LiveBench Data Analysis | 79.3% | – |
| LiveBench Language | 82.6% | – |
| LiveBench Instruction Following | 70.2% | – |
| AA Intelligence Index | 39.0 | 46.3 |
| IFBench | 74.0% | – |
| AA-LCR | 82.0% | 77.0% |
| MMMU-Pro | 78.4% | – |
| AA-Omniscience | 5.8 | 30.9 |
| Terminal-Bench Hard | 57.6% | – |
| GPQA Diamond (AA) | 92.0% | – |
| Humanity's Last Exam (AA) | 43.7% | 42.3% |
| SciCode (AA) | – | 57.8% |
| τ²-Bench Telecom (AA) | 87.1% | – |
| AIME (Vals) | 96.7% | – |
| LiveCodeBench | 84.1% | – |
| MMLU-Pro | 87.5% | – |
| LegalBench | 86.0% | 85.1% |
| CorpFin | 65.3% | – |
| TaxEval | 74.0% | – |
| MedQA | 96.1% | – |
| SWE-bench (Vals) | 78.2% | – |
| GPQA Diamond (Vals) | 91.7% | – |
| AIME 2026 | 99.2% | – |
| HMMT February 2026 | 97.7% | – |
| MathArena Apex | 54.2% | – |
| SWE-Bench Pro | 59.1% | – |
| MCP Atlas | 70.6% | – |
| ARC-AGI-1 | 93.7% | – |
| ARC-AGI-2 | 74.0% | – |
| CritPt | 23.4% | 18.0% |
| GDPval (AA) | 36.6% | 59.7% |
| τ²-Bench Banking (AA) | 39.6% | – |
| APEX-Agents (AA) | 33.3% | – |
| CaseLaw v2 | 63.8% | – |
| Code Migration | 35.0% | 34.7% |
| Excel Modeling Benchmark | – | 54.9% |
| Harvey's Legal Agent Benchmark | 0.0% | 19.6% |
| Legal Research Bench | – | 39.9% |
| MedCode | 41.3% | 48.7% |
| MedScribe | 77.5% | 87.2% |
| MMMU-Pro (Vals) | 87.5% | – |
| MortgageTax | 68.3% | – |
| ProgramBench | 0.0% | 0.0% |
| Public Benefits Bench | – | 68.5% |
| SAGE | 43.3% | 31.0% |
| SkillsBench | 51.7% | – |
| Vibe Code Bench v1.1 | 67.4% | – |
| SWE-Bench Pro (private) | 43.4% | – |
| SWE Atlas: Codebase QnA | 40.8% | – |
| SWE Atlas: Refactoring | 44.3% | – |
| SWE Atlas: Test Writing | 44.4% | – |
| Chess Puzzles | 44.0% | – |
| EBR-bench | 25.4% | – |
| Mystery Game Puzzles | 37.0% | – |
| CL-bench | 27.9% | – |
| CL-bench Life | 21.7% | – |
| METR Time Horizons | 74.3% | – |
| DeepSWE | 51.8% | – |
| LMCA | 52.0% | – |
| DTBench | 94.4% | – |
Data as of 2026-09-21. Best configuration of each model; every score links to its source on the model pages.
GPT-5.4 vs Grok 4.7: questions
- Is GPT-5.4 better than Grok 4.7?
- GPT-5.4 (xhigh) and Grok 4.7 (high) are level on quality (65.4 vs 65.8). The BenchLeader Index combines every independent quality benchmark; Grok 4.7 (high) is ahead overall as of 2026-09-21, but check the category scores for your use.
- Which is cheaper, GPT-5.4 or Grok 4.7?
- Grok 4.7 is cheaper: $3.00 against $5.63 per million tokens, blended at three input tokens per output token.
- Which is faster, GPT-5.4 or Grok 4.7?
- GPT-5.4 streams faster: 148 against 55 output tokens per second.
- Which has the larger context window?
- GPT-5.4 accepts more context: 1.1M against 500k tokens.