# DeepSeek V4 Flash 0731 vs Grok 4.20

> Side-by-side benchmark comparison for DeepSeek V4 Flash 0731 and Grok 4.20 across agentic, coding, multimodal, knowledge, reasoning, multilingual, and math tasks.

- Canonical page: https://benchlm.ai/compare/deepseek-v4-flash-0731-vs-grok-4-20-beta
- Last updated: September 18, 2026

- Shared sourced benchmarks: 9
- HTML indexing: indexable
- Ranking lane: BenchAlign v5

## Quick Verdict

Pick Grok 4.20 if you want the stronger benchmark profile. DeepSeek V4 Flash 0731 only makes more sense when its price, context window, or workload-specific category wins matter more than the overall score.

## Summary

- Grok 4.20 leads overall 67.27 to 56.
- The biggest single benchmark swing is Terminal-Bench 2.1 (Vals) in Agentic, with scores of 67.0% and 44.2%.
- DeepSeek V4 Flash 0731 is the cheaper option on output tokens, which matters if you expect large responses or heavy interactive use.
- Grok 4.20 also has the larger context window at 2M.

## Model Snapshot

| Property | DeepSeek V4 Flash 0731 | Grok 4.20 |
|----------|----------|----------|
| Creator | DeepSeek | xAI |
| Type | Proprietary | Proprietary |
| Reasoning | Reasoning | Reasoning |
| Context | 1M | 2M |
| Overall Score | null | 67.27 |
| Benchmarks Covered | 40 | 23 |
| Pricing (input/output) | $0.14 / $0.28 | $2.00 / $6.00 |

## Category Breakdown

### Agentic

- Winner: Coming soon
- DeepSeek V4 Flash 0731 public-lane score: Coming soon
- Grok 4.20 public-lane score: 26.5 (Supported · #149/154)

| Benchmark | DeepSeek V4 Flash 0731 | Grok 4.20 | Winner |
|-----------|-----------|-----------|--------|
| Terminal-Bench 2.0 | 56.9% | 47.1% | DeepSeek V4 Flash 0731 |
| BrowseComp | 73.2% | Coming soon | Coming soon |
| HLE w/ tools | 45.1% | Coming soon | Coming soon |
| MCP Atlas | 69% | Coming soon | Coming soon |
| Toolathlon | 47.8% | Coming soon | Coming soon |
| Terminal-Bench 2.1 | 82.7% | Coming soon | Coming soon |
| CyberGym | 76.7% | Coming soon | Coming soon |
| Toolathlon-Verified | 70.3% | Coming soon | Coming soon |
| Agents' Last Exam | 25.2% | Coming soon | Coming soon |
| AutomationBench | 25.1% | Coming soon | Coming soon |
| Terminal-Bench 2.1 (Vals) | 67.0% | 44.2% | DeepSeek V4 Flash 0731 |
| DeepSearchQA | Coming soon | 62.8% | Coming soon |
| Gert Labs | Coming soon | 38.36% | Coming soon |

### Coding

- Winner: Coming soon
- DeepSeek V4 Flash 0731 public-lane score: Coming soon
- Grok 4.20 public-lane score: 28.3 (Supported · #145/154)

| Benchmark | DeepSeek V4 Flash 0731 | Grok 4.20 | Winner |
|-----------|-----------|-----------|--------|
| LiveCodeBench Pass@1-COT | 91.6% | Coming soon | Coming soon |
| Codeforces | 3052.0 | Coming soon | Coming soon |
| SWE-bench Verified | 79% | 76.7% | DeepSeek V4 Flash 0731 |
| SWE-bench Pro | 52.6% | 51.8% | DeepSeek V4 Flash 0731 |
| SWE Multilingual | 73.3% | Coming soon | Coming soon |
| Terminal-Bench 2.0 | 56.9% | Coming soon | Coming soon |
| Terminal-Bench 2.1 | 82.7% | Coming soon | Coming soon |
| NL2Repo | 54.2% | Coming soon | Coming soon |
| DeepSWE | 54.4% | Coming soon | Coming soon |
| DSBench-FullStack | 68.7% | Coming soon | Coming soon |
| DSBench-Hard | 59.6% | Coming soon | Coming soon |
| VulcanBench v3 | 88.4% | Coming soon | Coming soon |
| OpenHarmony Bench | 53.8% | Coming soon | Coming soon |
| LiveCodeBench (Vals) | 87.3% | 84.3% | DeepSeek V4 Flash 0731 |
| SWE-bench (Vals) | 88.8% | 72.2% | DeepSeek V4 Flash 0731 |
| LiveCodeBench Pro | Coming soon | 74.2% | Coming soon |
| Vibe Code Bench | Coming soon | 4.06% | Coming soon |

### Multimodal & Grounded

- Winner: Coming soon
- DeepSeek V4 Flash 0731 public-lane score: Coming soon
- Grok 4.20 public-lane score: 34.6 (#43/48)

| Benchmark | DeepSeek V4 Flash 0731 | Grok 4.20 | Winner |
|-----------|-----------|-----------|--------|
| MMMU-Pro | Coming soon | 75.2% | Coming soon |
| CharXiv | Coming soon | 60.9% | Coming soon |
| ERQA | Coming soon | 54.1% | Coming soon |
| SimpleVQA | Coming soon | 57.4% | Coming soon |
| MedXpertQA (MM) | Coming soon | 65.8% | Coming soon |

### Reasoning

- Winner: Coming soon
- DeepSeek V4 Flash 0731 public-lane score: Coming soon
- Grok 4.20 public-lane score: 34.3 (Unranked · 2 rankable rows)

| Benchmark | DeepSeek V4 Flash 0731 | Grok 4.20 | Winner |
|-----------|-----------|-----------|--------|
| MRCR 1M | 78.7% | Coming soon | Coming soon |
| CorpusQA 1M | 60.5% | Coming soon | Coming soon |
| ARC-AGI-2 | Coming soon | 53.3% | Coming soon |
| ARC-AGI-3 | Coming soon | 0.1% | Coming soon |

### Knowledge

- Winner: Coming soon
- DeepSeek V4 Flash 0731 public-lane score: Coming soon
- Grok 4.20 public-lane score: 49.2 (Supported · #87/184)

| Benchmark | DeepSeek V4 Flash 0731 | Grok 4.20 | Winner |
|-----------|-----------|-----------|--------|
| MMLU-Pro | 86.2% | Coming soon | Coming soon |
| SimpleQA | 34.1% | Coming soon | Coming soon |
| Chinese-SimpleQA | 78.9% | Coming soon | Coming soon |
| GPQA | 88.1% | Coming soon | Coming soon |
| GPQA-D | 88.1% | 88.5% | Grok 4.20 |
| HLE | 34.8% | Coming soon | Coming soon |
| GPQA Diamond (Vals) | 89.9% | 88.6% | DeepSeek V4 Flash 0731 |
| MMLU-Pro (Vals) | 86.2% | 86.3% | Grok 4.20 |
| HLE w/o tools | Coming soon | 31.6% | Coming soon |
| HealthBench Hard | Coming soon | 20.3% | Coming soon |
| MedXpertQA (Text) | Coming soon | 50.2% | Coming soon |

### Mathematics

- Winner: Coming soon
- DeepSeek V4 Flash 0731 public-lane score: 79.9 (Unranked · 4 rankable rows)
- Grok 4.20 public-lane score: Coming soon

| Benchmark | DeepSeek V4 Flash 0731 | Grok 4.20 | Winner |
|-----------|-----------|-----------|--------|
| HMMT Feb 2026 | 94.8% | Coming soon | Coming soon |
| IMOAnswerBench | 88.4% | Coming soon | Coming soon |
| Apex | 33.0% | Coming soon | Coming soon |
| Apex Shortlist | 85.7% | Coming soon | Coming soon |

## FAQ

### Which is better overall, DeepSeek V4 Flash 0731 or Grok 4.20?

Grok 4.20 is ahead overall on BenchLM right now.

### How many benchmarks does DeepSeek V4 Flash 0731 cover on BenchLM?

DeepSeek V4 Flash 0731 currently has 40 sourced benchmark scores on BenchLM.

### How many benchmarks does Grok 4.20 cover on BenchLM?

Grok 4.20 currently has 23 sourced benchmark scores on BenchLM.

## Related Comparisons

- [DeepSeek V4 Flash 0731 vs Grok 4.20 Multi-agent](/compare/deepseek-v4-flash-0731-vs-grok-4-20-multi-agent-beta)
- [Grok 4.20 vs Grok 4.20 Multi-agent](/compare/grok-4-20-beta-vs-grok-4-20-multi-agent-beta)
- [DeepSeek V4 Flash 0731 vs Claude Fable 5.1](/compare/claude-fable-5-1-vs-deepseek-v4-flash-0731)
- [Grok 4.20 vs Claude Fable 5.1](/compare/claude-fable-5-1-vs-grok-4-20-beta)

## Explore More

- [DeepSeek V4 Flash 0731 profile](/models/deepseek-v4-flash-0731)
- [Grok 4.20 profile](/models/grok-4-20-beta)
- [Compare Pricing](/llm-pricing)
- [Alternative Finder](/tools/alternative-finder)
- [LLM Selector](/tools/llm-selector)
- [Overall Rankings](/best/overall)