# Grok 4.20 vs Grok 4.6

> Side-by-side benchmark comparison for Grok 4.20 and Grok 4.6 across agentic, coding, multimodal, knowledge, reasoning, multilingual, and math tasks.

- Canonical page: https://benchlm.ai/compare/grok-4-20-beta-vs-grok-4-6
- Last updated: September 25, 2026

- Shared sourced benchmarks: 7
- HTML indexing: indexable
- Ranking lane: BenchAlign v5.7

## Quick Verdict

Pick Grok 4.6 if you want the stronger benchmark profile. Grok 4.20 only makes more sense when its price, context window, or workload-specific category wins matter more than the overall score.

## Summary

- Grok 4.6 leads overall 69.19 to 59.66.
- The clearest category separation is in agentic, where the averages are 24.5 for Grok 4.20 and 67.9 for Grok 4.6.
- The biggest single benchmark swing is Terminal-Bench 2.1 (Vals) in Agentic, with scores of 44.2% and 78.3%.
- Grok 4.20 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 | Grok 4.20 | Grok 4.6 |
|----------|----------|----------|
| Creator | xAI | xAI |
| Type | Proprietary | Proprietary |
| Reasoning | Reasoning | Reasoning |
| Context | 2M | 500K |
| Overall Score | 59.66 | 69.19 |
| Benchmarks Covered | 23 | 17 |
| Pricing (input/output) | $1.25 / $2.50 | $2.00 / $6.00 |

## Category Breakdown

### Agentic

- Winner: Grok 4.6
- Grok 4.20 public-lane score: 24.5 (Supported · #81/105)
- Grok 4.6 public-lane score: 67.9 (Supported · #8/105)

| Benchmark | Grok 4.20 | Grok 4.6 | Winner |
|-----------|-----------|-----------|--------|
| Terminal-Bench 2.0 | 47.1% | Coming soon | Coming soon |
| DeepSearchQA | 62.8% | Coming soon | Coming soon |
| Gert Labs | 38.36% | Coming soon | Coming soon |
| Terminal-Bench 2.1 (Vals) | 44.2% | 78.3% | Grok 4.6 |
| Terminal-Bench 3.0 | Coming soon | 26.5% | Coming soon |
| APEX-Agents | Coming soon | 57.5% | Coming soon |
| ApprenticeBench | Coming soon | 13% | Coming soon |

### Coding

- Winner: Grok 4.6
- Grok 4.20 public-lane score: 23.8 (Supported · #106/135)
- Grok 4.6 public-lane score: 62 (Supported · #14/135)

| Benchmark | Grok 4.20 | Grok 4.6 | Winner |
|-----------|-----------|-----------|--------|
| LiveCodeBench Pro | 74.2% | Coming soon | Coming soon |
| SWE-bench Verified | 76.7% | Coming soon | Coming soon |
| SWE-bench Pro | 51.8% | Coming soon | Coming soon |
| Vibe Code Bench | 4.06% | Coming soon | Coming soon |
| LiveCodeBench (Vals) | 84.3% | 88.2% | Grok 4.6 |
| SWE-bench (Vals) | 72.2% | 95.6% | Grok 4.6 |
| Bug Hunt Bench | Coming soon | 27 fixes | Coming soon |
| DeepSWE | Coming soon | 65.9% | Coming soon |
| cursorBench32 | Coming soon | 70.8% | Coming soon |
| FrontierCode 1.1 Extended | Coming soon | 61.3% | Coming soon |
| VulcanBench v3 | Coming soon | 87.0% | Coming soon |
| FrontierSWE v2 | Coming soon | 25.3% | Coming soon |

### Reasoning

- Winner: Not comparable
- Grok 4.20 public-lane score: 41.1 (Unranked · 2 rankable rows)
- Grok 4.6 public-lane score: 57.6 (#16/19)

| Benchmark | Grok 4.20 | Grok 4.6 | Winner |
|-----------|-----------|-----------|--------|
| ARC-AGI-2 | 53.3% | 67.1% | Grok 4.6 |
| ARC-AGI-3 | 0.1% | 2.1% | Grok 4.6 |
| ARC-AGI-1 | Coming soon | 87.00% | Coming soon |

### Multimodal & Grounded

- Winner: Not comparable
- Grok 4.20 public-lane score: 34.6 (#45/50)
- Grok 4.6 public-lane score: Coming soon

| Benchmark | Grok 4.20 | Grok 4.6 | Winner |
|-----------|-----------|-----------|--------|
| MMMU-Pro | 75.2% | Coming soon | Coming soon |
| CharXiv | 60.9% | Coming soon | Coming soon |
| ERQA | 54.1% | Coming soon | Coming soon |
| SimpleVQA | 57.4% | Coming soon | Coming soon |
| MedXpertQA (MM) | 65.8% | Coming soon | Coming soon |

### Knowledge

- Winner: Grok 4.6
- Grok 4.20 public-lane score: 51 (Supported · #55/158)
- Grok 4.6 public-lane score: 69 (Supported · #13/158)

| Benchmark | Grok 4.20 | Grok 4.6 | Winner |
|-----------|-----------|-----------|--------|
| GPQA-D | 88.5% | Coming soon | Coming soon |
| HLE w/o tools | 31.6% | Coming soon | Coming soon |
| HealthBench Hard | 20.3% | Coming soon | Coming soon |
| MedXpertQA (Text) | 50.2% | Coming soon | Coming soon |
| GPQA Diamond (Vals) | 88.6% | 94.7% | Grok 4.6 |
| MMLU-Pro (Vals) | 86.3% | 89.4% | Grok 4.6 |

## FAQ

### Which is better overall, Grok 4.20 or Grok 4.6?

Grok 4.6 is ahead overall on BenchLM right now.

### Where is the biggest gap between Grok 4.20 and Grok 4.6?

The widest category gap is in agentic, where the averages are 24.5 for Grok 4.20 and 67.9 for Grok 4.6.

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

Grok 4.20 currently has 23 sourced benchmark scores on BenchLM.

### How many benchmarks does Grok 4.6 cover on BenchLM?

Grok 4.6 currently has 17 sourced benchmark scores on BenchLM.

## Related Comparisons

- [Grok 4.20 vs Grok 4.20 Multi-agent](/compare/grok-4-20-beta-vs-grok-4-20-multi-agent-beta)
- [Grok 4.6 vs Grok 4.20 Multi-agent](/compare/grok-4-20-multi-agent-beta-vs-grok-4-6)
- [Grok 4.20 vs GPT-6 Astra](/compare/gpt-6-astra-vs-grok-4-20-beta)
- [Grok 4.6 vs GPT-6 Astra](/compare/gpt-6-astra-vs-grok-4-6)

## Explore More

- [Grok 4.20 profile](/models/grok-4-20-beta)
- [Grok 4.6 profile](/models/grok-4-6)
- [Compare Pricing](/llm-pricing)
- [Alternative Finder](/tools/alternative-finder)
- [LLM Selector](/tools/llm-selector)
- [Overall Rankings](/best/overall)