# Grok 4.20 vs Grok 4.5

> Side-by-side benchmark comparison for Grok 4.20 and Grok 4.5 across agentic, coding, multimodal, knowledge, reasoning, multilingual, and math tasks.

- Canonical page: https://benchlm.ai/compare/grok-4-20-beta-vs-grok-4-5
- Last updated: October 7, 2026

- Shared sourced benchmarks: 8
- HTML indexing: indexable
- Ranking lane: BenchAlign v5.8

## Quick Verdict

Grok 4.5 leads by point estimate. Conditional score ranges do not establish rank confidence. Choose using the category evidence and your own workload tests.

## Summary

- Grok 4.5 has the higher overall point estimate 63.97 to 57.38. Conditional score ranges do not establish rank confidence.
- The clearest category separation is in multimodal & grounded, where the averages are 35.6 for Grok 4.20 and 79.5 for Grok 4.5.
- The biggest single benchmark swing is Terminal-Bench 2.1 (Vals) in Agentic, with scores of 44.2% and 67.8%.
- 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 4.5 |
|----------|----------|----------|
| Creator | xAI | xAI |
| Type | Proprietary | Proprietary |
| Reasoning | Reasoning | Reasoning |
| Context | 2M | 500K |
| Overall Score | 57.38 | 63.97 |
| Benchmarks Covered | 23 | 17 |
| Pricing (input/output) | $1.25 / $2.50 | $2.00 / $6.00 |

## Category Breakdown

### Agentic

- Winner: Grok 4.5
- Grok 4.20 public-lane score: 24.4 (Supported · #97/122)
- Grok 4.5 public-lane score: 57.6 (Supported · #29/122)

| Benchmark | Grok 4.20 | Grok 4.5 | 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% | 67.8% | Grok 4.5 |
| Terminal-Bench 3.0 | Coming soon | 15.7% | Coming soon |
| Terminal-Bench 2.1 | Coming soon | 83.3% | Coming soon |
| deepSwe | Coming soon | 53% | Coming soon |

### Coding

- Winner: Grok 4.5
- Grok 4.20 public-lane score: 23 (Supported · #114/146)
- Grok 4.5 public-lane score: 57.7 (Supported · #27/146)

| Benchmark | Grok 4.20 | Grok 4.5 | Winner |
|-----------|-----------|-----------|--------|
| LiveCodeBench Pro | 74.2% | Coming soon | Coming soon |
| SWE-bench Verified | 76.7% | Coming soon | Coming soon |
| SWE-bench Pro | 51.8% | 64.7% | Grok 4.5 |
| Vibe Code Bench | 4.06% | Coming soon | Coming soon |
| LiveCodeBench (Vals) | 84.3% | 87.4% | Grok 4.5 |
| SWE-bench (Vals) | 72.2% | 86.6% | Grok 4.5 |
| SWE Multilingual | Coming soon | 78% | Coming soon |
| Terminal-Bench 2.1 | Coming soon | 83.3% | Coming soon |
| CursorBench 3.2 | Coming soon | 66.7% | Coming soon |
| VulcanBench v3 | Coming soon | 89.9% | Coming soon |
| PostTrainBench v1.1 | Coming soon | 23.4% | Coming soon |

### Reasoning

- Winner: Not comparable
- Grok 4.20 public-lane score: 41.1 (Unranked · 2 rankable rows)
- Grok 4.5 public-lane score: 50.4 (#28/28)

| Benchmark | Grok 4.20 | Grok 4.5 | Winner |
|-----------|-----------|-----------|--------|
| ARC-AGI-2 | 53.3% | 52.6% | Grok 4.20 |
| ARC-AGI-3 | 0.1% | 0.3% | Grok 4.5 |
| ARC-AGI-1 | Coming soon | 85.67% | Coming soon |

### Multimodal & Grounded

- Winner: Not comparable
- Grok 4.20 public-lane score: 35.6 (#44/49)
- Grok 4.5 public-lane score: 79.5 (Unranked · 1 rankable row)

| Benchmark | Grok 4.20 | Grok 4.5 | 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 4.5
- Grok 4.20 public-lane score: 50.5 (Supported · #66/174)
- Grok 4.5 public-lane score: 67 (Supported · #19/174)

| Benchmark | Grok 4.20 | Grok 4.5 | 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% | 92.9% | Grok 4.5 |
| MMLU-Pro (Vals) | 86.3% | 89.2% | Grok 4.5 |

## FAQ

### Which is better overall, Grok 4.20 or Grok 4.5?

Grok 4.5 has the higher overall point estimate. Conditional score ranges do not establish rank confidence.

### Where is the biggest gap between Grok 4.20 and Grok 4.5?

The widest category gap is in multimodal & grounded, where the averages are 35.6 for Grok 4.20 and 79.5 for Grok 4.5.

### 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 4.5 cover on BenchLM?

Grok 4.5 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 4.5 vs Grok 4.20 Multi-agent](/compare/grok-4-20-multi-agent-beta-vs-grok-4-5)
- [Grok 4.20 vs Claude Opus 5.5](/compare/claude-opus-5-5-vs-grok-4-20-beta)
- [Grok 4.5 vs Claude Opus 5.5](/compare/claude-opus-5-5-vs-grok-4-5)

## Explore More

- [Grok 4.20 profile](/models/grok-4-20-beta)
- [Grok 4.5 profile](/models/grok-4-5)
- [Compare Pricing](/llm-pricing)
- [Alternative Finder](/tools/alternative-finder)
- [LLM Selector](/tools/llm-selector)
- [Overall Rankings](/best/overall)