Grok 4.6 vs Grok 4.7
Grok 4.6 Verdict
- Grok 4.6 (medium) and Grok 4.7 (high) are level on quality (66.0 vs 65.8).
- Grok 4.6 (medium) is stronger in agents & tools, coding, knowledge, long context.
- Grok 4.7 (high) is stronger in composite, reasoning.
- They cost about the same ($3.00 per 1M blended).
| Metric | Grok 4.6 (medium) | Grok 4.7 (high) |
|---|---|---|
| BenchLeader Index | 66.0 | 65.8 |
| Agents & tools score | 73.3 | – |
| Coding score | 64.1 | – |
| Composite score | 83.1 | 87.5 |
| Knowledge score | 77.3 | 72.6 |
| Long context score | 66.9 | 64.8 |
| Reasoning score | 67.6 | 76.4 |
| Blended price $/M | $3.00 | $3.00 |
| Output speed | 61 tok/s | 55 tok/s |
| Time to first answer | 34.9 s | 1.4 s |
| Context window | 500k | 500k |
| SciCode | 54.6% | – |
| AA Intelligence Index | 42.8 | 46.3 |
| AA-LCR | 81.0% | 77.0% |
| AA-Omniscience | 28 | 30.9 |
| GPQA Diamond (AA) | 93.5% | – |
| Humanity's Last Exam (AA) | 42.1% | 42.3% |
| SciCode (AA) | 55.9% | 57.8% |
| LegalBench | – | 85.1% |
| ARC-AGI-1 | 87.5% | – |
| ARC-AGI-2 | 61.3% | – |
| CritPt | 17.7% | 18.0% |
| GDPval (AA) | 55.2% | 59.7% |
| τ²-Bench Banking (AA) | 44.3% | – |
| Code Migration | – | 34.7% |
| Excel Modeling Benchmark | – | 54.9% |
| Harvey's Legal Agent Benchmark | – | 19.6% |
| Legal Research Bench | – | 39.9% |
| MedCode | – | 48.7% |
| MedScribe | – | 87.2% |
| ProgramBench | – | 0.0% |
| Public Benefits Bench | – | 68.5% |
| SAGE | – | 31.0% |
| DeepSWE | 67.5% | – |
| CursorBench | 67.1% | – |
Data as of 2026-09-21. Best configuration of each model; every score links to its source on the model pages.
Grok 4.6 vs Grok 4.7: questions
- Is Grok 4.6 better than Grok 4.7?
- Grok 4.6 (medium) and Grok 4.7 (high) are level on quality (66.0 vs 65.8). The BenchLeader Index combines every independent quality benchmark; Grok 4.6 (medium) is ahead overall as of 2026-09-21, but check the category scores for your use.
- Which is cheaper, Grok 4.6 or Grok 4.7?
- Grok 4.6 is cheaper: $3.00 against $3.00 per million tokens, blended at three input tokens per output token.
- Which is faster, Grok 4.6 or Grok 4.7?
- Grok 4.6 streams faster: 61 against 55 output tokens per second.
- Which has the larger context window?
- Both accept 500k tokens of context.