# GPT-5 vs GPT-5.3 Codex

> GPT-5 is the stronger model overall, scoring 50.9 to 45.8 on the Noometry Index.

- Canonical page: https://noometry.com/compare/gpt-5-vs-gpt-5-3-codex
- Last updated: 2026-10-11
- Shared benchmarks: 7

## Summary

- They share 7 benchmarks with published results for both. GPT-5 scores higher in 1 category and GPT-5.3 Codex in 1 category; 2 gaps are clear of the uncertainty.
- The widest gap is in agentic & tool use, where GPT-5.3 Codex leads 48.0 to 33.1.
- The biggest single-benchmark swing is Terminal-Bench: 49.6% for GPT-5 and 78.4% for GPT-5.3 Codex.
- GPT-5 is cheaper at $1.25 / $10 per million input/output tokens, against $1.75 / $14 for GPT-5.3 Codex.

## Snapshot

| | GPT-5 | GPT-5.3 Codex |
|---|---|---|
| Provider | OpenAI | OpenAI |
| Noometry Index | 50.9 | 45.8 |
| Rank | 45 | 69 |
| Context | 400K | 400K |
| Input $/M | $1.25 | $1.75 |
| Output $/M | $10 | $14 |
| Weights | Proprietary | Proprietary |

## Coding

- GPT-5: 50.3 (#47)
- GPT-5.3 Codex: 48.6 (#56)

| Benchmark | GPT-5 | GPT-5.3 Codex |
|---|---|---|
| SWE-bench Verified | 73.6% | 74.8% |
| LMArena WebDev | 1418 | 1409 |
| WeirdML | 60.7% | 79.3% |
| ALE-Bench | 1,162 | 1,655 |
| SWE-bench Verified (bash only) | 65% | — |
| Aider Polyglot | 88% | — |
| SciCode | 42.9% | — |
| GSO | 6.9% | — |
| LMArena Coding | 1436 | — |
| AlgoTune | 1.67 | — |

## Agentic & Tool Use

- GPT-5: 33.1 (#56)
- GPT-5.3 Codex: 48.0 (#9)

| Benchmark | GPT-5 | GPT-5.3 Codex |
|---|---|---|
| Terminal-Bench | 49.6% | 78.4% |
| METR Time Horizons | 69.6% | 74.5% |
| GDPval | 34.8% | — |
| Remote Labor Index | 1.7% | — |
| DeepResearch Bench | 49.6% | — |
| BALROG | 32.8% | — |
| LMArena Search | 1133 | — |
| Vending-Bench 2 | — | 5,940 |

## Reasoning

- GPT-5: 38.3 (#64)
- GPT-5.3 Codex: —

| Benchmark | GPT-5 | GPT-5.3 Codex |
|---|---|---|
| Epoch Capabilities Index | 150 | 156.77 |
| ARC-AGI-2 | 9.9% | — |
| SimpleBench | 56.7% | — |
| Kagi LLM Benchmark | 72.7% | — |
| ARC-AGI-1 | 65.7% | — |
| CritPt | 12.6% | — |
| Chess Puzzles | 37% | — |
| EnigmaEval | 10.5% | — |
| EBR-Bench | 12.7% | — |
| LMArena Hard Prompts | 1416 | — |
| Mystery Game Puzzles | 23% | — |
| DTBench | 90.7% | — |
| LMCA | 40% | — |
| ForecastBench | 61.4 | — |

## Math

- GPT-5: 55.0 (#44)
- GPT-5.3 Codex: —

| Benchmark | GPT-5 | GPT-5.3 Codex |
|---|---|---|
| FrontierMath (Tiers 1-3) | 55.4% | — |
| FrontierMath Tier 4 | 22% | — |
| OTIS Mock AIME 2024-2025 | 91.4% | — |
| ProofBench | 18% | — |
| Omni-MATH | 64.7% | — |
| LMArena Math | 1407 | — |
| MATH Level 5 | 98.1% | — |
| FrontierMath (Feb 2025 set) | 32.4% | — |
| FrontierMath Tier 4 (v1) | 12.5% | — |

## Knowledge

- GPT-5: 56.6 (#43)
- GPT-5.3 Codex: —

| Benchmark | GPT-5 | GPT-5.3 Codex |
|---|---|---|
| GPQA Diamond | 86.2% | — |
| Humanity's Last Exam | 25.3% | — |
| SimpleQA Verified | 50.1% | — |
| MMLU-Pro | 86.3% | — |
| Confabulations | 10.3% | — |
| Vectara Hallucination Rate | 14.7% | — |
| GPQA (HELM) | 79.2% | — |
| LMArena Expert | 1419 | — |

## Multimodal

- GPT-5: 46.8 (#13)
- GPT-5.3 Codex: —

| Benchmark | GPT-5 | GPT-5.3 Codex |
|---|---|---|
| LMArena Vision | 1232 | — |
| GeoBench | 81% | — |
| VPCT | 66% | — |

## Multilingual

- GPT-5: 51.4 (#110)
- GPT-5.3 Codex: —

| Benchmark | GPT-5 | GPT-5.3 Codex |
|---|---|---|
| LMArena Non-English | 1397 | — |
| LMArena Chinese | 1422 | — |
| LMArena French | 1410 | — |
| LMArena German | 1416 | — |
| LMArena Japanese | 1409 | — |
| LMArena Korean | 1360 | — |
| LMArena Russian | 1406 | — |
| LMArena Spanish | 1399 | — |

## Instruction Following

- GPT-5: 73.8 (#113)
- GPT-5.3 Codex: —

| Benchmark | GPT-5 | GPT-5.3 Codex |
|---|---|---|
| IFEval | 87.5% | — |
| LMArena Instruction Following | 1388 | — |

## Long Context

- GPT-5: 69.5 (#2)
- GPT-5.3 Codex: —

| Benchmark | GPT-5 | GPT-5.3 Codex |
|---|---|---|
| Fiction.LiveBench | 97.2% | — |
| LMArena Longer Query | 1399 | — |

## Writing & Preference

- GPT-5: 63.4 (#65)
- GPT-5.3 Codex: —

| Benchmark | GPT-5 | GPT-5.3 Codex |
|---|---|---|
| LMArena Text | 1406 | — |
| LMArena Creative Writing | 1365 | — |
| Short-Story Creative Writing | 86% | — |
| EQ-Bench Creative Writing | 1627 | — |
| WildBench | 85.7% | — |
| LMArena Multi-Turn | 1426 | — |

## FAQ

### Is GPT-5 better than GPT-5.3 Codex?

GPT-5 is the stronger model overall, scoring 50.9 to 45.8 on the Noometry Index.

### Which is cheaper, GPT-5 or GPT-5.3 Codex?

GPT-5 is cheaper. It lists at $1.25 per million input tokens and $10 per million output tokens; GPT-5.3 Codex lists at $1.75 and $14.

### Is GPT-5 or GPT-5.3 Codex better for coding?

GPT-5 scores higher on coding benchmarks: 50.3 versus 48.6 in the Noometry coding category.

### Which has the bigger context window?

Both accept 400K tokens.

### How many benchmarks do GPT-5 and GPT-5.3 Codex share?

7 benchmarks have published results for both models. GPT-5 has 69 scored results on Noometry and GPT-5.3 Codex has 8.