Model comparison
GPT-5.2 Pro vs GPT-5.4 mini
GPT-5.2 Pro is the stronger model overall, scoring 52.2 to 44.8 on the Noometry Index. GPT-5.4 mini costs 34× less per token, which makes it the better buy when GPT-5.2 Pro's lead doesn't matter for your workload.
Last verified . 7 shared benchmarks.
Summary
- They share 7 benchmarks with published results for both. GPT-5.2 Pro scores higher in 2 categories and GPT-5.4 mini in 0 categories; 2 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where GPT-5.2 Pro leads 51.8 to 30.3.
- The biggest single-benchmark swing is FrontierMath Tier 4: 46% for GPT-5.2 Pro and 9.8% for GPT-5.4 mini.
- GPT-5.4 mini is cheaper at $0.75 / $4.50 per million input/output tokens, against $21 / $168 for GPT-5.2 Pro.
Side by side
| GPT-5.2 Pro | GPT-5.4 mini | |
|---|---|---|
| Provider | OpenAI | OpenAI |
| Noometry Index | 52.2 | 44.8 |
| Released | 2025-12-11 | 2026-03-17 |
| Weights | Proprietary | Proprietary |
| Context window | 400K | 400K |
| Max output | 128K | 128K |
| Input $ / M tokens | $21 | $0.75 |
| Output $ / M tokens | $168 | $4.50 |
| Results tracked | 8 | 46 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Not comparable
GPT-5.2 Pro: —, GPT-5.4 mini: 45.0 (#67)
| Benchmark | GPT-5.2 Pro | GPT-5.4 mini |
|---|---|---|
| FrontierCode | — | 27% |
| LMArena WebDev | — | 1397 |
| SciCode | — | 49.9% |
| WeirdML | — | 60.3% |
| LMArena Coding | — | 1438 |
| ALE-Bench | — | 1,189 |
Agentic & Tool Use Not comparable
GPT-5.2 Pro: —, GPT-5.4 mini: 29.6 (#74)
| Benchmark | GPT-5.2 Pro | GPT-5.4 mini |
|---|---|---|
| DeepResearch Bench | — | 36.3% |
Reasoning GPT-5.2 Pro leads
GPT-5.2 Pro: 51.8 (#31), GPT-5.4 mini: 30.3 (#77)
| Benchmark | GPT-5.2 Pro | GPT-5.4 mini |
|---|---|---|
| ARC-AGI-2 | 54.2% | 18.9% |
| NYT Connections (extended) | 79.3% | 61.8% |
| ARC-AGI-1 | 90.5% | 63.7% |
| Epoch Capabilities Index | 155.41 | 148.83 |
| SimpleBench | 57.4% | — |
| Kagi LLM Benchmark | — | 37.9% |
| CritPt | — | 10% |
| Chess Puzzles | — | 24% |
| Thematic Generalization | — | 61.7% |
| LMArena Hard Prompts | — | 1424 |
| Mystery Game Puzzles | — | 11% |
| DTBench | — | 80% |
| LMCA | — | 40.8% |
| ForecastBench | — | 57 |
Math GPT-5.2 Pro leads
GPT-5.2 Pro: 64.8 (#23), GPT-5.4 mini: 45.4 (#63)
| Benchmark | GPT-5.2 Pro | GPT-5.4 mini |
|---|---|---|
| FrontierMath (Tiers 1-3) | 74% | 51.2% |
| FrontierMath Tier 4 | 46% | 9.8% |
| FrontierMath Tier 4 (v1) | 31.3% | 2.1% |
| OTIS Mock AIME 2024-2025 | — | 88.9% |
| ProofBench | — | 21% |
| LMArena Math | — | 1418 |
| FrontierMath (Feb 2025 set) | — | 28.3% |
Knowledge Not comparable
GPT-5.2 Pro: —, GPT-5.4 mini: 51.4 (#62)
| Benchmark | GPT-5.2 Pro | GPT-5.4 mini |
|---|---|---|
| GPQA Diamond | — | 86.9% |
| SimpleQA Verified | — | 29.4% |
| Vectara Hallucination Rate | — | 5.5% |
| LMArena Expert | — | 1434 |
Multimodal Not comparable
GPT-5.2 Pro: —, GPT-5.4 mini: 39.0 (#51)
| Benchmark | GPT-5.2 Pro | GPT-5.4 mini |
|---|---|---|
| LMArena Vision | — | 1245 |
Multilingual Not comparable
GPT-5.2 Pro: —, GPT-5.4 mini: 52.1 (#94)
| Benchmark | GPT-5.2 Pro | GPT-5.4 mini |
|---|---|---|
| LMArena Non-English | — | 1405 |
| LMArena Chinese | — | 1446 |
| LMArena French | — | 1440 |
| LMArena German | — | 1407 |
| LMArena Japanese | — | 1375 |
| LMArena Korean | — | 1369 |
| LMArena Russian | — | 1417 |
| LMArena Spanish | — | 1403 |
Instruction Following Not comparable
GPT-5.2 Pro: —, GPT-5.4 mini: 74.3 (#99)
| Benchmark | GPT-5.2 Pro | GPT-5.4 mini |
|---|---|---|
| LMArena Instruction Following | — | 1405 |
Long Context Not comparable
GPT-5.2 Pro: —, GPT-5.4 mini: 42.9 (#107)
| Benchmark | GPT-5.2 Pro | GPT-5.4 mini |
|---|---|---|
| LMArena Longer Query | — | 1407 |
Writing & Preference Not comparable
GPT-5.2 Pro: —, GPT-5.4 mini: 64.2 (#56)
| Benchmark | GPT-5.2 Pro | GPT-5.4 mini |
|---|---|---|
| LMArena Text | — | 1411 |
| LMArena Creative Writing | — | 1370 |
| EQ-Bench Creative Writing | — | 1665 |
| LMArena Multi-Turn | — | 1429 |
Frequently asked questions
Is GPT-5.2 Pro better than GPT-5.4 mini?
GPT-5.2 Pro is the stronger model overall, scoring 52.2 to 44.8 on the Noometry Index. GPT-5.4 mini costs 34× less per token, which makes it the better buy when GPT-5.2 Pro's lead doesn't matter for your workload.
Which is cheaper, GPT-5.2 Pro or GPT-5.4 mini?
GPT-5.4 mini is cheaper. It lists at $0.75 per million input tokens and $4.50 per million output tokens; GPT-5.2 Pro lists at $21 and $168.
Which has the bigger context window?
Both accept 400K tokens.
How many benchmarks do GPT-5.2 Pro and GPT-5.4 mini share?
7 benchmarks have published results for both models. GPT-5.2 Pro has 8 scored results on Noometry and GPT-5.4 mini has 46.