Model comparison
GPT-5.4 mini vs Mercury 2
GPT-5.4 mini is the stronger model overall, scoring 44.8 to 38.8 on the Noometry Index. Mercury 2 costs 4.5× less per token, which makes it the better buy when GPT-5.4 mini's lead doesn't matter for your workload.
Last verified . 17 shared benchmarks.
Summary
- They share 17 benchmarks with published results for both. GPT-5.4 mini scores higher in 7 categories and Mercury 2 in 0 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where GPT-5.4 mini leads 51.4 to 35.6.
- The biggest single-benchmark swing is WeirdML: 60.3% for GPT-5.4 mini and 43.2% for Mercury 2.
- Mercury 2 is cheaper at $0.25 / $0.75 per million input/output tokens, against $0.75 / $4.50 for GPT-5.4 mini.
- GPT-5.4 mini accepts more context: 400K tokens versus 128K.
Side by side
| GPT-5.4 mini | Mercury 2 | |
|---|---|---|
| Provider | OpenAI | Inception |
| Noometry Index | 44.8 | 38.8 |
| Released | 2026-03-17 | 2026-02-20 |
| Weights | Proprietary | Proprietary |
| Context window | 400K | 128K |
| Max output | 128K | 50K |
| Input $ / M tokens | $0.75 | $0.25 |
| Output $ / M tokens | $4.50 | $0.75 |
| Results tracked | 46 | 17 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5.4 mini leads
GPT-5.4 mini: 45.0 (#67), Mercury 2: 33.2 (#250)
| Benchmark | GPT-5.4 mini | Mercury 2 |
|---|---|---|
| LMArena WebDev | 1397 | 1170 |
| SciCode | 49.9% | 38.7% |
| WeirdML | 60.3% | 43.2% |
| LMArena Coding | 1438 | 1391 |
| ALE-Bench | 1,189 | 785.58 |
| FrontierCode | 27% | — |
Agentic & Tool Use Not comparable
GPT-5.4 mini: 29.6 (#74), Mercury 2: —
| Benchmark | GPT-5.4 mini | Mercury 2 |
|---|---|---|
| DeepResearch Bench | 36.3% | — |
Reasoning GPT-5.4 mini leads
GPT-5.4 mini: 30.3 (#77), Mercury 2: 23.0 (#170)
| Benchmark | GPT-5.4 mini | Mercury 2 |
|---|---|---|
| CritPt | 10% | 0.8% |
| LMArena Hard Prompts | 1424 | 1362 |
| ARC-AGI-2 | 18.9% | — |
| Kagi LLM Benchmark | 37.9% | — |
| NYT Connections (extended) | 61.8% | — |
| ARC-AGI-1 | 63.7% | — |
| Chess Puzzles | 24% | — |
| Thematic Generalization | 61.7% | — |
| Mystery Game Puzzles | 11% | — |
| DTBench | 80% | — |
| LMCA | 40.8% | — |
| Epoch Capabilities Index | 148.83 | — |
| ForecastBench | 57 | — |
Math Not comparable
GPT-5.4 mini: 45.4 (#63), Mercury 2: —
| Benchmark | GPT-5.4 mini | Mercury 2 |
|---|---|---|
| FrontierMath (Tiers 1-3) | 51.2% | — |
| FrontierMath Tier 4 | 9.8% | — |
| OTIS Mock AIME 2024-2025 | 88.9% | — |
| ProofBench | 21% | — |
| LMArena Math | 1418 | — |
| FrontierMath (Feb 2025 set) | 28.3% | — |
| FrontierMath Tier 4 (v1) | 2.1% | — |
Knowledge GPT-5.4 mini leads
GPT-5.4 mini: 51.4 (#62), Mercury 2: 35.6 (#170)
| Benchmark | GPT-5.4 mini | Mercury 2 |
|---|---|---|
| Vectara Hallucination Rate | 5.5% | 12.3% |
| LMArena Expert | 1434 | 1358 |
| GPQA Diamond | 86.9% | — |
| SimpleQA Verified | 29.4% | — |
Multimodal Not comparable
GPT-5.4 mini: 39.0 (#51), Mercury 2: —
| Benchmark | GPT-5.4 mini | Mercury 2 |
|---|---|---|
| LMArena Vision | 1245 | — |
Multilingual GPT-5.4 mini leads
GPT-5.4 mini: 52.1 (#94), Mercury 2: 46.8 (#153)
| Benchmark | GPT-5.4 mini | Mercury 2 |
|---|---|---|
| LMArena Non-English | 1405 | 1331 |
| LMArena Chinese | 1446 | 1416 |
| LMArena Russian | 1417 | 1303 |
| LMArena French | 1440 | — |
| LMArena German | 1407 | — |
| LMArena Japanese | 1375 | — |
| LMArena Korean | 1369 | — |
| LMArena Spanish | 1403 | — |
Instruction Following GPT-5.4 mini leads
GPT-5.4 mini: 74.3 (#99), Mercury 2: 70.4 (#161)
| Benchmark | GPT-5.4 mini | Mercury 2 |
|---|---|---|
| LMArena Instruction Following | 1405 | 1328 |
Long Context GPT-5.4 mini leads
GPT-5.4 mini: 42.9 (#107), Mercury 2: 40.4 (#149)
| Benchmark | GPT-5.4 mini | Mercury 2 |
|---|---|---|
| LMArena Longer Query | 1407 | 1329 |
Writing & Preference GPT-5.4 mini leads
GPT-5.4 mini: 64.2 (#56), Mercury 2: 54.0 (#151)
| Benchmark | GPT-5.4 mini | Mercury 2 |
|---|---|---|
| LMArena Text | 1411 | 1355 |
| LMArena Creative Writing | 1370 | 1289 |
| LMArena Multi-Turn | 1429 | 1358 |
| EQ-Bench Creative Writing | 1665 | — |
Frequently asked questions
Is GPT-5.4 mini better than Mercury 2?
GPT-5.4 mini is the stronger model overall, scoring 44.8 to 38.8 on the Noometry Index. Mercury 2 costs 4.5× less per token, which makes it the better buy when GPT-5.4 mini's lead doesn't matter for your workload.
Which is cheaper, GPT-5.4 mini or Mercury 2?
Mercury 2 is cheaper. It lists at $0.25 per million input tokens and $0.75 per million output tokens; GPT-5.4 mini lists at $0.75 and $4.50.
Is GPT-5.4 mini or Mercury 2 better for coding?
GPT-5.4 mini scores higher on coding benchmarks: 45.0 versus 33.2 in the Noometry coding category.
Which has the bigger context window?
GPT-5.4 mini does, with 400K tokens against 128K.
How many benchmarks do GPT-5.4 mini and Mercury 2 share?
17 benchmarks have published results for both models. GPT-5.4 mini has 46 scored results on Noometry and Mercury 2 has 17.