token.app › Compare › GPT-5 vs Grok 4.6
GPT-5 vs Grok 4.6
Grok 4.6 leads on 4 of 4 shared benchmarks. Grok 4.6 is 1.1× cheaper per token on a blended 3:1 basis. Grok 4.6 accepts the larger context window (500K).
Head to head
Blended price uses a 3:1 input:output mix
| GPT-5 | Grok 4.6 | |
|---|---|---|
| Provider | OpenAI | SpaceXAI |
| Input $/1M | $1.25 | $2.00 |
| Output $/1M | $10.00 | $6.00 |
| Blended $/1M | $3.44 | $3.00 |
| Context window | 400K | 500K |
| Max output | 128K | 450K |
| Released | Aug 2025 | Aug 2026 |
| Licence | Proprietary | Proprietary |
| GPQA Diamond | 86.2% | 94.0% |
| FrontierMath T1–3 | 55.4% | 66.0% |
| FrontierMath T4 | 22.0% | 31.7% |
| OTIS Mock AIME | 91.4% | 99.2% |
Benchmark scores independently run and published by Epoch AI, 'AI Benchmarking Hub', used under CC BY 4.0. Only benchmarks with a score for BOTH models are compared.
Choose GPT-5 if…
- No clear advantage on the data we hold.
Choose Grok 4.6 if…
- Costs less per token — $3.00 vs $3.44 blended
- Higher GPQA Diamond (94.0% vs 86.2%)
- Higher FrontierMath T1–3 (66.0% vs 55.4%)
- Higher FrontierMath T4 (31.7% vs 22.0%)
- Higher OTIS Mock AIME (99.2% vs 91.4%)
- Larger context window (500K)
- Newer model (released Aug 2026)
FAQ
Which is better, GPT-5 or Grok 4.6?
Across the 4 benchmarks both models have independently-run scores for, GPT-5 leads on 0 and Grok 4.6 leads on 4. Benchmarks are one input — price and context window are on this page too.
Is GPT-5 cheaper than Grok 4.6?
Grok 4.6 is cheaper: $3.00 versus $3.44 per million tokens blended at 3:1 input:output. Input alone: $1.25 vs $2.00. Output alone: $10.00 vs $6.00.
What context windows do GPT-5 and Grok 4.6 have?
GPT-5: 400,000 tokens. Grok 4.6: 500,000 tokens.
Who makes GPT-5 and Grok 4.6?
GPT-5 is made by OpenAI. Grok 4.6 is made by SpaceXAI.