token.app › Compare › Grok 4.20 vs DeepSeek V4 Pro
Grok 4.20 vs DeepSeek V4 Pro
DeepSeek V4 Pro leads on 4 of 5 shared benchmarks. DeepSeek V4 Pro is 2.9× 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 | DeepSeek V4 Pro | |
|---|---|---|
| Provider | xAI | DeepSeek |
| Input $/1M | $1.25 | $0.435 |
| Output $/1M | $2.50 | $0.870 |
| Blended $/1M | $1.56 | $0.544 |
| Context window | 2M | 1.0M |
| Max output | — | 384K |
| Released | Mar 2026 | Apr 2026 |
| Licence | Proprietary | Open weights |
| GPQA Diamond | 89.3% | 90.9% |
| FrontierMath T1–3 | 44.9% | 45.3% |
| FrontierMath T4 | 17.1% | 2.4% |
| SimpleQA Verified | 35.1% | 57.0% |
| OTIS Mock AIME | 92.2% | 96.7% |
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…
- Higher FrontierMath T4 (17.1% vs 2.4%)
- Larger context window (2M)
Choose DeepSeek V4 Pro if…
- Costs less per token — $0.544 vs $1.56 blended
- Higher GPQA Diamond (90.9% vs 89.3%)
- Higher FrontierMath T1–3 (45.3% vs 44.9%)
- Higher SimpleQA Verified (57.0% vs 35.1%)
- Higher OTIS Mock AIME (96.7% vs 92.2%)
- Open weights — self-hostable
- Newer model (released Apr 2026)
FAQ
Which is better, Grok 4.20 or DeepSeek V4 Pro?
Across the 5 benchmarks both models have independently-run scores for, Grok 4.20 leads on 1 and DeepSeek V4 Pro leads on 4. Benchmarks are one input — price and context window are on this page too.
Is Grok 4.20 cheaper than DeepSeek V4 Pro?
DeepSeek V4 Pro is cheaper: $0.544 versus $1.56 per million tokens blended at 3:1 input:output. Input alone: $1.25 vs $0.435. Output alone: $2.50 vs $0.870.
What context windows do Grok 4.20 and DeepSeek V4 Pro have?
Grok 4.20: 2,000,000 tokens. DeepSeek V4 Pro: 1,048,576 tokens.
Who makes Grok 4.20 and DeepSeek V4 Pro?
Grok 4.20 is made by xAI. DeepSeek V4 Pro is made by DeepSeek.