token.app › Compare › Grok 4.20 vs GPT-5.6 Luna
Grok 4.20 vs GPT-5.6 Luna
GPT-5.6 Luna leads on 5 of 5 shared benchmarks. GPT-5.6 Luna is 6.9× cheaper per token on a blended 3:1 basis. Grok 4.20 accepts the larger context window (2M).
Head to head
Blended price uses a 3:1 input:output mix
| Grok 4.20 | GPT-5.6 Luna | |
|---|---|---|
| Provider | xAI | OpenAI |
| Input $/1M | $1.25 | $0.100 |
| Output $/1M | $2.50 | $0.600 |
| Blended $/1M | $1.56 | $0.225 |
| Context window | 2M | 1.1M |
| Max output | — | 128K |
| Released | Mar 2026 | Jul 2026 |
| Licence | Proprietary | Proprietary |
| GPQA Diamond | 89.3% | 91.6% |
| FrontierMath T1–3 | 44.9% | 82.1% |
| FrontierMath T4 | 17.1% | 61.0% |
| SimpleQA Verified | 35.1% | 41.7% |
| OTIS Mock AIME | 92.2% | 98.3% |
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 Grok 4.20 if…
- Larger context window (2M)
Choose GPT-5.6 Luna if…
- Costs less per token — $0.225 vs $1.56 blended
- Higher GPQA Diamond (91.6% vs 89.3%)
- Higher FrontierMath T1–3 (82.1% vs 44.9%)
- Higher FrontierMath T4 (61.0% vs 17.1%)
- Higher SimpleQA Verified (41.7% vs 35.1%)
- Newer model (released Jul 2026)
FAQ
Which is better, Grok 4.20 or GPT-5.6 Luna?
Across the 5 benchmarks both models have independently-run scores for, Grok 4.20 leads on 0 and GPT-5.6 Luna leads on 5. Benchmarks are one input — price and context window are on this page too.
Is Grok 4.20 cheaper than GPT-5.6 Luna?
GPT-5.6 Luna is cheaper: $0.225 versus $1.56 per million tokens blended at 3:1 input:output. Input alone: $1.25 vs $0.100. Output alone: $2.50 vs $0.600.
What context windows do Grok 4.20 and GPT-5.6 Luna have?
Grok 4.20: 2,000,000 tokens. GPT-5.6 Luna: 1,050,000 tokens.
Who makes Grok 4.20 and GPT-5.6 Luna?
Grok 4.20 is made by xAI. GPT-5.6 Luna is made by OpenAI.