# GPT-4.1 vs GPT-5.2

> Side-by-side benchmark comparison for GPT-4.1 and GPT-5.2 across agentic, coding, multimodal, knowledge, reasoning, multilingual, and math tasks.

- Canonical page: https://benchlm.ai/compare/gpt-4-1-vs-gpt-5-2
- Last updated: September 4, 2026

- Shared sourced benchmarks: 5
- HTML indexing: indexable
- Ranking lane: BenchAlign v5

## Quick Verdict

Pick GPT-5.2 if you want the stronger benchmark profile. GPT-4.1 only makes more sense when its price, context window, or workload-specific category wins matter more than the overall score.

## Summary

- GPT-5.2 leads overall 65.45 to 43.81.
- The clearest category separation is in instruction following, where the averages are 50.3 for GPT-4.1 and 92.3 for GPT-5.2.
- The biggest single benchmark swing is FrontierMath v2 (Tiers 1-3) in Mathematics, with scores of 5.517% and 40.700%.
- GPT-4.1 is the cheaper option on output tokens, which matters if you expect large responses or heavy interactive use.
- GPT-4.1 also has the larger context window at 1M.

## Model Snapshot

| Property | GPT-4.1 | GPT-5.2 |
|----------|----------|----------|
| Creator | OpenAI | OpenAI |
| Type | Proprietary | Proprietary |
| Reasoning | Non-Reasoning | Reasoning |
| Context | 1M | 400K |
| Overall Score | 43.81 | 65.45 |
| Benchmarks Covered | 7 | 15 |
| Pricing (input/output) | $2.00 / $8.00 | $1.75 / $14.00 |

## Category Breakdown

### Agentic

- Winner: Coming soon
- GPT-4.1 public-lane score: Coming soon
- GPT-5.2 public-lane score: 43 (Supported · #108/151)

| Benchmark | GPT-4.1 | GPT-5.2 | Winner |
|-----------|-----------|-----------|--------|
| Gert Labs | 25.65% | 46.54% | GPT-5.2 |
| BrowseComp | Coming soon | 65.8% | Coming soon |
| OSWorld-Verified | Coming soon | 47.3% | Coming soon |
| JobBench | Coming soon | 34.3% | Coming soon |

### Coding

- Winner: GPT-5.2
- GPT-4.1 public-lane score: 38.8 (Supported · #143/183)
- GPT-5.2 public-lane score: 47.3 (Supported · #91/183)

| Benchmark | GPT-4.1 | GPT-5.2 | Winner |
|-----------|-----------|-----------|--------|
| SWE-bench Verified | 54.6% | 80% | GPT-5.2 |
| SWE-bench Pro | Coming soon | 55.6% | Coming soon |
| Vibe Code Bench | Coming soon | 53.50% | Coming soon |

### Multimodal & Grounded

- Winner: Coming soon
- GPT-4.1 public-lane score: 50.2 (Unranked · 1 rankable row)
- GPT-5.2 public-lane score: 66.3 (#23/48)

| Benchmark | GPT-4.1 | GPT-5.2 | Winner |
|-----------|-----------|-----------|--------|
| MMMU-Pro | Coming soon | 79.5% | Coming soon |
| MathVision | Coming soon | 83.0% | Coming soon |
| CharXiv | Coming soon | 82.1% | Coming soon |
| V* | Coming soon | 75.9% | Coming soon |

### Reasoning

- Winner: Coming soon
- GPT-4.1 public-lane score: 67.2 (Unranked · 2 rankable rows)
- GPT-5.2 public-lane score: 53.8 (Unranked · 3 rankable rows)

| Benchmark | GPT-4.1 | GPT-5.2 | Winner |
|-----------|-----------|-----------|--------|
| ARC-AGI-2 | Coming soon | 52.9% | Coming soon |

### Knowledge

- Winner: GPT-5.2
- GPT-4.1 public-lane score: 40.7 (Supported · #135/181)
- GPT-5.2 public-lane score: 62 (Supported · #30/181)

| Benchmark | GPT-4.1 | GPT-5.2 | Winner |
|-----------|-----------|-----------|--------|
| MMLU | 90.2% | Coming soon | Coming soon |
| GPQA | 66.3% | 92.4% | GPT-5.2 |

### Instruction Following

- Winner: Coming soon
- GPT-4.1 public-lane score: 50.3 (#80/120)
- GPT-5.2 public-lane score: 92.3 (#14/120)

| Benchmark | GPT-4.1 | GPT-5.2 | Winner |
|-----------|-----------|-----------|--------|
| IFEval | 87.4% | Coming soon | Coming soon |

### Mathematics

- Winner: Coming soon
- GPT-4.1 public-lane score: 28.1 (Unranked · 2 rankable rows)
- GPT-5.2 public-lane score: 57.5 (Unranked · 2 rankable rows)

| Benchmark | GPT-4.1 | GPT-5.2 | Winner |
|-----------|-----------|-----------|--------|
| FrontierMath v2 (Tiers 1-3) | 5.517% | 40.700% | GPT-5.2 |
| FrontierMath v2 (Tier 4) | 0.000% | 18.800% | GPT-5.2 |

## FAQ

### Which is better overall, GPT-4.1 or GPT-5.2?

GPT-5.2 is ahead overall on BenchLM right now.

### Where is the biggest gap between GPT-4.1 and GPT-5.2?

The widest category gap is in instruction following, where the averages are 50.3 for GPT-4.1 and 92.3 for GPT-5.2.

### How many benchmarks does GPT-4.1 cover on BenchLM?

GPT-4.1 currently has 7 sourced benchmark scores on BenchLM.

### How many benchmarks does GPT-5.2 cover on BenchLM?

GPT-5.2 currently has 15 sourced benchmark scores on BenchLM.

## Related Comparisons

- [GPT-4.1 vs GPT-4.1 mini](/compare/gpt-4-1-vs-gpt-4-1-mini)
- [GPT-5.2 vs GPT-4.1 mini](/compare/gpt-4-1-mini-vs-gpt-5-2)
- [GPT-4.1 vs GPT-4.1 nano](/compare/gpt-4-1-vs-gpt-4-1-nano)
- [GPT-5.2 vs GPT-4.1 nano](/compare/gpt-4-1-nano-vs-gpt-5-2)

## Explore More

- [GPT-4.1 profile](/models/gpt-4-1)
- [GPT-5.2 profile](/models/gpt-5-2)
- [Compare Pricing](/llm-pricing)
- [Alternative Finder](/tools/alternative-finder)
- [LLM Selector](/tools/llm-selector)
- [Overall Rankings](/best/overall)