# GPT-5.4 nano vs Mercury 2

> GPT-5.4 nano is the stronger model overall, scoring 41.9 to 39.1 on the Noometry Index.

- Canonical page: https://noometry.com/compare/gpt-5-4-nano-vs-mercury-2
- Last updated: 2026-10-12
- Shared benchmarks: 16

## Summary

- They share 16 benchmarks with published results for both. GPT-5.4 nano scores higher in 6 categories and Mercury 2 in 1 category; 6 gaps are clear of the uncertainty.
- The widest gap is in coding, where GPT-5.4 nano leads 43.6 to 33.5.
- The biggest single-benchmark swing is Vectara Hallucination Rate: 3.1% for GPT-5.4 nano and 12.3% for Mercury 2.
- Mercury 2 is cheaper at $0.25 / $0.75 per million input/output tokens, against $0.20 / $1.25 for GPT-5.4 nano.
- GPT-5.4 nano accepts more context: 400K tokens versus 128K.

## Snapshot

| | GPT-5.4 nano | Mercury 2 |
|---|---|---|
| Provider | OpenAI | Inception |
| Noometry Index | 41.9 | 39.1 |
| Rank | 125 | 175 |
| Context | 400K | 128K |
| Input $/M | $0.20 | $0.25 |
| Output $/M | $1.25 | $0.75 |
| Weights | Proprietary | Proprietary |

## Coding

- GPT-5.4 nano: 43.6 (#84)
- Mercury 2: 33.5 (#255)

| Benchmark | GPT-5.4 nano | Mercury 2 |
|---|---|---|
| SciCode | 46.9% | 38.7% |
| WeirdML | 49.2% | 43.2% |
| LMArena Coding | 1405 | 1391 |
| ALE-Bench | 1,005 | 785.58 |
| LMArena WebDev | — | 1171 |

## Reasoning

- GPT-5.4 nano: 23.7 (#173)
- Mercury 2: 23.8 (#170)

| Benchmark | GPT-5.4 nano | Mercury 2 |
|---|---|---|
| CritPt | 9.3% | 0.8% |
| LMArena Hard Prompts | 1381 | 1362 |
| ARC-AGI-2 | 5.7% | — |
| Kagi LLM Benchmark | 39.7% | — |
| ARC-AGI-1 | 51.5% | — |
| Chess Puzzles | 30% | — |
| Mystery Game Puzzles | 9% | — |
| DTBench | 80.3% | — |
| LMCA | 36.9% | — |
| Epoch Capabilities Index | 145.81 | — |
| ForecastBench | 57.3 | — |

## Math

- GPT-5.4 nano: 40.9 (#88)
- Mercury 2: —

| Benchmark | GPT-5.4 nano | Mercury 2 |
|---|---|---|
| FrontierMath (Tiers 1-3) | 44.9% | — |
| FrontierMath Tier 4 | 12.2% | — |
| OTIS Mock AIME 2024-2025 | 87.8% | — |
| ProofBench | 5% | — |
| LMArena Math | 1406 | — |
| FrontierMath (Feb 2025 set) | 25.9% | — |
| FrontierMath Tier 4 (v1) | 6.3% | — |

## Knowledge

- GPT-5.4 nano: 41.9 (#103)
- Mercury 2: 36.2 (#172)

| Benchmark | GPT-5.4 nano | Mercury 2 |
|---|---|---|
| Vectara Hallucination Rate | 3.1% | 12.3% |
| LMArena Expert | 1396 | 1358 |
| GPQA Diamond | 78.5% | — |
| SimpleQA Verified | 11.7% | — |

## Multimodal

- GPT-5.4 nano: 36.7 (#78)
- Mercury 2: —

| Benchmark | GPT-5.4 nano | Mercury 2 |
|---|---|---|
| LMArena Vision | 1196 | — |

## Multilingual

- GPT-5.4 nano: 48.6 (#140)
- Mercury 2: 46.6 (#157)

| Benchmark | GPT-5.4 nano | Mercury 2 |
|---|---|---|
| LMArena Non-English | 1359 | 1331 |
| LMArena Chinese | 1392 | 1417 |
| LMArena Russian | 1363 | 1304 |
| LMArena French | 1396 | — |
| LMArena German | 1367 | — |
| LMArena Japanese | 1343 | — |
| LMArena Korean | 1320 | — |
| LMArena Spanish | 1371 | — |

## Instruction Following

- GPT-5.4 nano: 71.9 (#144)
- Mercury 2: 70.2 (#165)

| Benchmark | GPT-5.4 nano | Mercury 2 |
|---|---|---|
| LMArena Instruction Following | 1362 | 1329 |

## Long Context

- GPT-5.4 nano: 41.6 (#137)
- Mercury 2: 40.5 (#154)

| Benchmark | GPT-5.4 nano | Mercury 2 |
|---|---|---|
| LMArena Longer Query | 1366 | 1330 |

## Writing & Preference

- GPT-5.4 nano: 55.7 (#142)
- Mercury 2: 53.8 (#155)

| Benchmark | GPT-5.4 nano | Mercury 2 |
|---|---|---|
| LMArena Text | 1372 | 1355 |
| LMArena Creative Writing | 1314 | 1289 |
| LMArena Multi-Turn | 1382 | 1358 |

## FAQ

### Is GPT-5.4 nano better than Mercury 2?

GPT-5.4 nano is the stronger model overall, scoring 41.9 to 39.1 on the Noometry Index.

### Which is cheaper, GPT-5.4 nano 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 nano lists at $0.20 and $1.25.

### Is GPT-5.4 nano or Mercury 2 better for coding?

GPT-5.4 nano scores higher on coding benchmarks: 43.6 versus 33.5 in the Noometry coding category.

### Which has the bigger context window?

GPT-5.4 nano does, with 400K tokens against 128K.

### How many benchmarks do GPT-5.4 nano and Mercury 2 share?

16 benchmarks have published results for both models. GPT-5.4 nano has 40 scored results on Noometry and Mercury 2 has 17.