Model comparison
GPT-5.3 Codex vs Mercury 2
GPT-5.3 Codex is the stronger model overall, scoring 45.8 to 39.1 on the Noometry Index. Mercury 2 costs 13× less per token, which makes it the better buy when GPT-5.3 Codex's lead doesn't matter for your workload.
Last verified . 3 shared benchmarks.
Summary
- They share 3 benchmarks with published results for both. GPT-5.3 Codex scores higher in 1 category and Mercury 2 in 0 categories; one gap is clear of the uncertainty.
- The widest gap is in coding, where GPT-5.3 Codex leads 48.6 to 33.5.
- The biggest single-benchmark swing is WeirdML: 79.3% for GPT-5.3 Codex and 43.2% for Mercury 2.
- Mercury 2 is cheaper at $0.25 / $0.75 per million input/output tokens, against $1.75 / $14 for GPT-5.3 Codex.
- GPT-5.3 Codex accepts more context: 400K tokens versus 128K.
Side by side
| GPT-5.3 Codex | Mercury 2 | |
|---|---|---|
| Provider | OpenAI | Inception |
| Noometry Index | 45.8 | 39.1 |
| Released | 2026-02-05 | 2026-02-20 |
| Weights | Proprietary | Proprietary |
| Context window | 400K | 128K |
| Max output | 128K | 50K |
| Input $ / M tokens | $1.75 | $0.25 |
| Output $ / M tokens | $14 | $0.75 |
| Results tracked | 8 | 17 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5.3 Codex leads
GPT-5.3 Codex: 48.6 (#56), Mercury 2: 33.5 (#255)
| Benchmark | GPT-5.3 Codex | Mercury 2 |
|---|---|---|
| LMArena WebDev | 1409 | 1171 |
| WeirdML | 79.3% | 43.2% |
| ALE-Bench | 1,655 | 785.58 |
| SWE-bench Verified | 74.8% | — |
| SciCode | — | 38.7% |
| LMArena Coding | — | 1391 |
Agentic & Tool Use Not comparable
GPT-5.3 Codex: 48.0 (#9), Mercury 2: —
| Benchmark | GPT-5.3 Codex | Mercury 2 |
|---|---|---|
| Terminal-Bench | 78.4% | — |
| METR Time Horizons | 74.5% | — |
| Vending-Bench 2 | 5,940 | — |
Reasoning Not comparable
GPT-5.3 Codex: —, Mercury 2: 23.8 (#170)
| Benchmark | GPT-5.3 Codex | Mercury 2 |
|---|---|---|
| CritPt | — | 0.8% |
| LMArena Hard Prompts | — | 1362 |
| Epoch Capabilities Index | 156.77 | — |
Knowledge Not comparable
GPT-5.3 Codex: —, Mercury 2: 36.2 (#172)
| Benchmark | GPT-5.3 Codex | Mercury 2 |
|---|---|---|
| Vectara Hallucination Rate | — | 12.3% |
| LMArena Expert | — | 1358 |
Multilingual Not comparable
GPT-5.3 Codex: —, Mercury 2: 46.6 (#157)
| Benchmark | GPT-5.3 Codex | Mercury 2 |
|---|---|---|
| LMArena Non-English | — | 1331 |
| LMArena Chinese | — | 1417 |
| LMArena Russian | — | 1304 |
Instruction Following Not comparable
GPT-5.3 Codex: —, Mercury 2: 70.2 (#165)
| Benchmark | GPT-5.3 Codex | Mercury 2 |
|---|---|---|
| LMArena Instruction Following | — | 1329 |
Long Context Not comparable
GPT-5.3 Codex: —, Mercury 2: 40.5 (#154)
| Benchmark | GPT-5.3 Codex | Mercury 2 |
|---|---|---|
| LMArena Longer Query | — | 1330 |
Writing & Preference Not comparable
GPT-5.3 Codex: —, Mercury 2: 53.8 (#155)
| Benchmark | GPT-5.3 Codex | Mercury 2 |
|---|---|---|
| LMArena Text | — | 1355 |
| LMArena Creative Writing | — | 1289 |
| LMArena Multi-Turn | — | 1358 |
Frequently asked questions
Is GPT-5.3 Codex better than Mercury 2?
GPT-5.3 Codex is the stronger model overall, scoring 45.8 to 39.1 on the Noometry Index. Mercury 2 costs 13× less per token, which makes it the better buy when GPT-5.3 Codex's lead doesn't matter for your workload.
Which is cheaper, GPT-5.3 Codex or Mercury 2?
Mercury 2 is cheaper. It lists at $0.25 per million input tokens and $0.75 per million output tokens; GPT-5.3 Codex lists at $1.75 and $14.
Is GPT-5.3 Codex or Mercury 2 better for coding?
GPT-5.3 Codex scores higher on coding benchmarks: 48.6 versus 33.5 in the Noometry coding category.
Which has the bigger context window?
GPT-5.3 Codex does, with 400K tokens against 128K.
How many benchmarks do GPT-5.3 Codex and Mercury 2 share?
3 benchmarks have published results for both models. GPT-5.3 Codex has 8 scored results on Noometry and Mercury 2 has 17.