# GPT-5.3 Codex vs Mistral Medium

> GPT-5.3 Codex is the stronger model overall, scoring 45.5 to 36.0 on the Noometry Index. Mistral Medium costs 1.6× less per token, which makes it the better buy when GPT-5.3 Codex's lead doesn't matter for your workload.

- Canonical page: https://noometry.com/compare/gpt-5-3-codex-vs-mistral-medium
- Last updated: 2026-10-10
- Shared benchmarks: 2

## Summary

- They share 2 benchmarks with published results for both. GPT-5.3 Codex scores higher in 2 categories and Mistral Medium in 0 categories; 2 gaps are clear of the uncertainty.
- The widest gap is in agentic & tool use, where GPT-5.3 Codex leads 47.8 to 28.1.
- The biggest single-benchmark swing is WeirdML: 79.3% for GPT-5.3 Codex and 43.7% for Mistral Medium.
- Mistral Medium is cheaper at $1.50 / $7.50 per million input/output tokens, against $1.75 / $14 for GPT-5.3 Codex.
- GPT-5.3 Codex accepts more context: 400K tokens versus 262K.
- Mistral Medium has downloadable open weights; the other is API-only.

## Snapshot

| | GPT-5.3 Codex | Mistral Medium |
|---|---|---|
| Provider | OpenAI | Mistral AI |
| Noometry Index | 45.5 | 36.0 |
| Rank | 61 | 211 |
| Context | 400K | 262K |
| Input $/M | $1.75 | $1.50 |
| Output $/M | $14 | $7.50 |
| Weights | Proprietary | Open |

## Coding

- GPT-5.3 Codex: 48.5 (#51)
- Mistral Medium: 33.8 (#241)

| Benchmark | GPT-5.3 Codex | Mistral Medium |
|---|---|---|
| WeirdML | 79.3% | 43.7% |
| ALE-Bench | 1,655 | 763.98 |
| SWE-bench Verified | 74.8% | — |
| FrontierCode | — | 8% |
| LMArena WebDev | 1408 | — |
| SciCode | — | 40.2% |
| LMArena Coding | — | 1434 |

## Agentic & Tool Use

- GPT-5.3 Codex: 47.8 (#9)
- Mistral Medium: 28.1 (#84)

| Benchmark | GPT-5.3 Codex | Mistral Medium |
|---|---|---|
| Terminal-Bench | 78.4% | — |
| Berkeley Function Calling Leaderboard | — | 37.7% |
| METR Time Horizons | 74.5% | — |
| Vending-Bench 2 | 5,940 | — |

## Reasoning

- GPT-5.3 Codex: —
- Mistral Medium: 23.3 (#165)

| Benchmark | GPT-5.3 Codex | Mistral Medium |
|---|---|---|
| Kagi LLM Benchmark | — | 50% |
| CritPt | — | 0% |
| LMArena Hard Prompts | — | 1426 |
| DTBench | — | 75.5% |
| LMCA | — | 26.1% |
| Surface Evolver Bench | — | 26.9% |
| Epoch Capabilities Index | 156.84 | — |

## Math

- GPT-5.3 Codex: —
- Mistral Medium: 27.2 (#238)

| Benchmark | GPT-5.3 Codex | Mistral Medium |
|---|---|---|
| OTIS Mock AIME 2024-2025 | — | 32.2% |
| ProofBench | — | 9% |
| LMArena Math | — | 1409 |
| MATH Level 5 | — | 81.6% |
| FrontierMath (Feb 2025 set) | — | 0.3% |

## Knowledge

- GPT-5.3 Codex: —
- Mistral Medium: 24.9 (#260)

| Benchmark | GPT-5.3 Codex | Mistral Medium |
|---|---|---|
| GPQA Diamond | — | 59.5% |
| Humanity's Last Exam | — | 4.5% |
| Vectara Hallucination Rate | — | 22.7% |
| LMArena Expert | — | 1408 |

## Multimodal

- GPT-5.3 Codex: —
- Mistral Medium: 34.6 (#82)

| Benchmark | GPT-5.3 Codex | Mistral Medium |
|---|---|---|
| LMArena Vision | — | 1172 |

## Multilingual

- GPT-5.3 Codex: —
- Mistral Medium: 52.3 (#86)

| Benchmark | GPT-5.3 Codex | Mistral Medium |
|---|---|---|
| LMArena Non-English | — | 1408 |
| LMArena Chinese | — | 1448 |
| LMArena French | — | 1460 |
| LMArena German | — | 1433 |
| LMArena Japanese | — | 1378 |
| LMArena Korean | — | 1382 |
| LMArena Russian | — | 1411 |
| LMArena Spanish | — | 1433 |

## Instruction Following

- GPT-5.3 Codex: —
- Mistral Medium: 73.9 (#111)

| Benchmark | GPT-5.3 Codex | Mistral Medium |
|---|---|---|
| LMArena Instruction Following | — | 1398 |

## Long Context

- GPT-5.3 Codex: —
- Mistral Medium: 42.9 (#108)

| Benchmark | GPT-5.3 Codex | Mistral Medium |
|---|---|---|
| LMArena Longer Query | — | 1406 |

## Writing & Preference

- GPT-5.3 Codex: —
- Mistral Medium: 60.2 (#97)

| Benchmark | GPT-5.3 Codex | Mistral Medium |
|---|---|---|
| LMArena Text | — | 1424 |
| LMArena Creative Writing | — | 1391 |
| Short-Story Creative Writing | — | 77.3% |
| LMArena Multi-Turn | — | 1418 |

## FAQ

### Is GPT-5.3 Codex better than Mistral Medium?

GPT-5.3 Codex is the stronger model overall, scoring 45.5 to 36.0 on the Noometry Index. Mistral Medium costs 1.6× 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 Mistral Medium?

Mistral Medium is cheaper. It lists at $1.50 per million input tokens and $7.50 per million output tokens; GPT-5.3 Codex lists at $1.75 and $14.

### Is GPT-5.3 Codex or Mistral Medium better for coding?

GPT-5.3 Codex scores higher on coding benchmarks: 48.5 versus 33.8 in the Noometry coding category.

### Which has the bigger context window?

GPT-5.3 Codex does, with 400K tokens against 262K.

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

2 benchmarks have published results for both models. GPT-5.3 Codex has 8 scored results on Noometry and Mistral Medium has 36.