DeepSeek-V4-Pro-0813 vs Grok 4.20
DeepSeek-V4-Pro-0813 (DeepSeek) and Grok 4.20 (xAI) compared on benchmarks, pricing, context window, and use-case rankings.
- DeepSeek-V4-Pro-0813 leads on overall intelligence (ECI 155 vs 152).
- DeepSeek-V4-Pro-0813 is cheaper on input tokens — $0.58 vs $1.25 per 1M.
- Grok 4.20 has a larger context window (2M vs 1M tokens).
- DeepSeek-V4-Pro-0813 ranks higher for coding (76th vs 42th percentile).
| Attribute | DeepSeek-V4-Pro-0813 | Grok 4.20 |
|---|---|---|
| Scores | ||
| Intelligence (ECI) | 155 | 152 |
| Coding | 76 | 42 |
| Math | 88 | 58 |
| Reasoning & Knowledge | 69 | 47 |
| Agentic & Tools | 67 | 55 |
| Specifications | ||
| Developer | DeepSeek | xAI |
| Family | DeepSeek | Grok |
| Released | Aug 13, 2026 | Feb 17, 2026 |
| Parameters | 1.6T | 500B |
| Availability | Open weights (unrestricted) | API access |
| Context window | 1M | 2M |
| Price — $/M input | $0.58 | $1.25 |
| Price — $/M output | $1.74 | $2.50 |
| Inputs | text | text, image, file |
| Outputs | text | text |
| Benchmarks | ||
| AIME 2024/2025 | 99% | 92% |
| APEX | 47% | — |
| ARC-AGI | 91% | 90% |
| ARC-AGI-2 | 61% | 65% |
| GPQA Diamond | 92% | 89% |
| Humanity's Last Exam | 41% | — |
| SciCode | 51% | — |
| SimpleQA Verified | 53% | 30% |
| WebDev Arena | 1581 | 1374 |
| WeirdML | 66% | 52% |
| Terminal-Bench | — | 57% |
Use-case scores are 0–100 percentile composites across each area’s benchmarks, ranked against every model from the past year. Highlighted cells lead each row. Open a model for the full picture.
Frequently asked questions
Is DeepSeek-V4-Pro-0813 better than Grok 4.20?
On Epoch AI's Capabilities Index, DeepSeek-V4-Pro-0813 scores higher (155) than Grok 4.20 (152). The right pick depends on your task — compare their coding, math, and reasoning scores in the table above.
Which is cheaper, DeepSeek-V4-Pro-0813 or Grok 4.20?
DeepSeek-V4-Pro-0813 is cheaper on input tokens at $0.58 per million, versus $1.25 (representative OpenRouter pricing).
Which has a larger context window, DeepSeek-V4-Pro-0813 or Grok 4.20?
Grok 4.20 supports up to 2M tokens, compared with 1M for the other.
Which is better for coding, DeepSeek-V4-Pro-0813 or Grok 4.20?
Across coding benchmarks like SWE-bench Verified and Terminal-Bench, DeepSeek-V4-Pro-0813 ranks higher — 76th vs 42th percentile among the models tracked on Model Beat.
Want a different match-up? Open the compare tool to add or swap models.
More comparisons
Benchmarks & model data from Epoch AI (CC BY); pricing & specs from OpenRouter. ECI = Epoch Capabilities Index.