Model comparison
GPT-4.1 vs GPT-5.4
Head-to-head evidence from 17 shared benchmark results across 7 categories. Overall scores shown here use the public BenchAlign v5 ranking lane.
Public leaderboard positions: GPT-4.1 #108 (Supported); GPT-5.4 #8 (Supported). Intervals and evidence labels describe ranking uncertainty, not a guarantee for a specific workload.
Evidence parity. GPT-4.1 and GPT-5.4 share 17 comparable benchmark results. 3 of 8 categories are comparable. 3 results are unique to GPT-4.1; 35 to GPT-5.4.
Updated July 22, 2026- Shared results
- 17
- GPT-4.1 only
- 3
- GPT-5.4 only
- 35
- Comparable categories
- 3 / 8
Pick GPT-5.4 if you want the stronger benchmark profile. GPT-4.1 only becomes the better choice if knowledge is the priority or you want the cheaper token bill.
Confidence note. This is a partial-evidence comparison with 17 shared benchmark results across 7 evidence categories; 3 of 8 categories currently have scoreable aggregates for both models. Treat the verdict as directional until coverage is more balanced.
Why this result
GPT-5.4 is clearly ahead on the BenchAlign aggregate, 74.24 to 51.11. The gap is large enough that you do not need to squint at the spreadsheet to see the difference.
GPT-5.4's sharpest advantage is in mathematics, where it averages 42.5 against 4.1. The single biggest benchmark swing on the page is FrontierMath v2 (Tiers 1-3), 5.517% to 47.600%. GPT-4.1 does hit back in knowledge, so the answer changes if that is the part of the workload you care about most.
GPT-5.4 is also the more expensive model on tokens at $2.50 input / $15.00 output per 1M tokens, versus $2.00 input / $8.00 output per 1M tokens for GPT-4.1. GPT-5.4 is the reasoning model in the pair, while GPT-4.1 is not. That usually helps on harder chain-of-thought-heavy tests, but it can also mean more latency and more token spend in real use. GPT-5.4 gives you the larger context window at 1.05M, compared with 1M for GPT-4.1.
Category breakdown
Exact category averages are shown below. Not measured means BenchLM does not have enough sourced public coverage for that model and category.
| Category | GPT-4.1 | Δ | GPT-5.4 |
|---|---|---|---|
| Math | GPT-4.14.1 | Margin→ 38.4 | GPT-5.442.5 |
| Knowledge | GPT-4.166.3 | Margin← 8.7 | GPT-5.457.6 |
| Coding | GPT-4.154.6 | Margin→ 3.1 | GPT-5.457.7 |
| Agentic | GPT-4.1Not measured | MarginNo overlap | GPT-5.477.2 |
| Multimodal | GPT-4.1Not measured | MarginNo overlap | GPT-5.473.2 |
| Inst. Following | GPT-4.187.4 | MarginNo overlap | GPT-5.4Not measured |
Decisive benchmark drivers
The largest measured benchmark gaps in this matchup, with exact reported values.
More
- Source ↗
FrontierMath v2 (Tiers 1-3)
MathA 5.517%B 47.600%Winner: GPT-5.4Δ 42.1FrontierMath v2 (Tiers 1-3): GPT-4.1 scored 5.517%; GPT-5.4 scored 47.600%. GPT-5.4 wins this benchmark. - Source ↗
FrontierMath v2 (Tier 4)
MathA 0.000%B 27.100%Winner: GPT-5.4Δ 27.1FrontierMath v2 (Tier 4): GPT-4.1 scored 0.000%; GPT-5.4 scored 27.100%. GPT-5.4 wins this benchmark. - Source ↗
GPQA
KnowledgeA 66.3%B 92.8%Winner: GPT-5.4Δ 26.5GPQA: GPT-4.1 scored 66.3%; GPT-5.4 scored 92.8%. GPT-5.4 wins this benchmark.
Operational comparison
Runtime and commercial metrics are compared only when both models have a complete sourced value.
| Metric | GPT-4.1 | GPT-5.4 | Comparison |
|---|---|---|---|
| Input / output priceUSD per 1M tokens | GPT-4.1$2 input / $8 output | GPT-5.4$2.5 input / $15 output | GPT-4.1 has the lower combined listed price. |
| Generation speedtokens per second | GPT-4.1108 tok/s | GPT-5.474 tok/s | GPT-4.1 has the higher measured throughput. |
| First-answer latencyseconds to first token | GPT-4.11.02 s | GPT-5.4151.79 s | GPT-4.1 reaches the first token sooner. |
| Context windowmaximum listed tokens | GPT-4.11M | GPT-5.41.05M | GPT-5.4 lists the larger context window. |
Benchmark Deep Dive
Agentic17 benchmarks
| Benchmark | GPT-4.1 | GPT-5.4 | Result |
|---|---|---|---|
| τ²-bench resultsSource | 47.1% | 87.1% | GPT-5.4 leads |
| Gert LabsSource | 25.65% | 64.89% | GPT-5.4 leads |
| Terminal-Bench 2.0Source | — | 75.1% | Not comparable |
| CyberGymSource | — | 79.0% | Not comparable |
| BrowseCompSource | — | 82.7% | Not comparable |
| OSWorld-VerifiedSource | — | 75% | Not comparable |
| MCP AtlasSource | — | 70.6% | Not comparable |
| ToolathlonSource | — | 54.6% | Not comparable |
| Claw-EvalSource | — | 60.3% | Not comparable |
| DeepSearchQASource | — | 73.6% | Not comparable |
| AA Agentic IndexSource | — | 41.1% | Not comparable |
| APEX-Agents-AASource | — | 33.3% | Not comparable |
| GDPval-AASource | — | 44.7% | Not comparable |
| GDPval-AASource | — | 1395 | Not comparable |
| ResearchClawBenchSource | — | 15.3% | Not comparable |
| JobBenchSource | — | 38.9% | Not comparable |
| ExploitGymSource | — | 6.0% | Not comparable |
CodingGPT-5.4 wins7 benchmarks
| Benchmark | GPT-4.1 | GPT-5.4 | Result |
|---|---|---|---|
| SWE-bench VerifiedSource | 54.6% | — | Not comparable |
| AA-SciCodeSource | 38.1% | 56.6% | GPT-5.4 leads |
| LiveCodeBench ProSource | — | 87.5% | Not comparable |
| SWE-bench ProSource | — | 57.7% | Not comparable |
| React Native EvalsSource | — | 85.3% | Not comparable |
| Vibe Code BenchSource | — | 67.42% | Not comparable |
| AA Coding IndexSource | — | 71.0% | Not comparable |
Reasoning2 benchmarks
KnowledgeGPT-4.1 wins14 benchmarks
| Benchmark | GPT-4.1 | GPT-5.4 | Result |
|---|---|---|---|
| MMLUSource | 90.2% | — | Not comparable |
| GPQASource | 66.3% | 92.8% | GPT-5.4 leads |
| Artificial Analysis Intelligence IndexSource | 19.4% | 51.4% | GPT-5.4 leads |
| AA-GPQA DiamondSource | 66.6% | 92.0% | GPT-5.4 leads |
| AA-HLESource | 4.6% | 41.6% | GPT-5.4 leads |
| AA-Omniscience IndexSource | -36.2% | 5.7% | GPT-5.4 leads |
| AA-Omniscience AccuracySource | 24.2% | 50.0% | GPT-5.4 leads |
| AA-Omniscience Hallucination RateSource | 79.6% | 88.6% | GPT-4.1 leads |
| HLESource | — | 52.1% | Not comparable |
| HLE w/o toolsSource | — | 39.8% | Not comparable |
| GPQA-DSource | — | 92.8% | Not comparable |
| HealthBench HardSource | — | 40.1% | Not comparable |
| MedXpertQA (Text)Source | — | 59.6% | Not comparable |
| HealthBench ProfessionalSource | — | 48.1% | Not comparable |
MathGPT-5.4 wins2 benchmarks
Multimodal11 benchmarks
| Benchmark | GPT-4.1 | GPT-5.4 | Result |
|---|---|---|---|
| AA-MMMU-ProSource | 61.2% | 78.4% | GPT-5.4 leads |
| Design Arena WebsiteSource | 1068 | 1250 | GPT-5.4 leads |
| MMMU-ProSource | — | 81.2% | Not comparable |
| OfficeQA ProSource | — | 53.2% | Not comparable |
| MMMU-Pro w/ PythonSource | — | 82.1% | Not comparable |
| CharXivSource | — | 82.8% | Not comparable |
| ERQASource | — | 65.4% | Not comparable |
| SimpleVQASource | — | 61.1% | Not comparable |
| ScreenSpot ProSource | — | 85.4% | Not comparable |
| ZeroBenchSource | — | 41.0% | Not comparable |
| MedXpertQA (MM)Source | — | 77.1% | Not comparable |
Frequently Asked Questions (4)
Which is better, GPT-4.1 or GPT-5.4?
GPT-5.4 is ahead on BenchLM's BenchAlign leaderboard, 74.24 to 51.11. The biggest single separator in this matchup is FrontierMath v2 (Tiers 1-3), where the scores are 5.517% and 47.600%.
Which is better for knowledge tasks, GPT-4.1 or GPT-5.4?
GPT-4.1 has the edge for knowledge tasks in this comparison, averaging 66.3 versus 57.6. Inside this category, AA-Omniscience Index is the benchmark that creates the most daylight between them.
Which is better for coding, GPT-4.1 or GPT-5.4?
GPT-5.4 has the edge for coding in this comparison, averaging 57.7 versus 54.6. Inside this category, AA-SciCode is the benchmark that creates the most daylight between them.
Which is better for math, GPT-4.1 or GPT-5.4?
GPT-5.4 has the edge for math in this comparison, averaging 42.5 versus 4.1. Inside this category, FrontierMath v2 (Tiers 1-3) is the benchmark that creates the most daylight between them.
Related Comparisons
Explore More
Choose a model with this week’s evidence
Join 2,000+ readers for ranking moves, pricing changes, and the claims that still need proof.
One email each week. Unsubscribe anytime.