# GPT-5.3 Codex vs Kimi K2.6

> Kimi K2.6 is the stronger model overall, scoring 47.5 to 45.5 on the Noometry Index.

- Canonical page: https://noometry.com/compare/gpt-5-3-codex-vs-kimi-k2-6
- Last updated: 2026-10-10
- Shared benchmarks: 6

## Summary

- They share 6 benchmarks with published results for both. GPT-5.3 Codex scores higher in 1 category and Kimi K2.6 in 1 category; 2 gaps are clear of the uncertainty.
- The widest gap is in agentic & tool use, where GPT-5.3 Codex leads 47.8 to 21.8.
- The biggest single-benchmark swing is WeirdML: 79.3% for GPT-5.3 Codex and 55.9% for Kimi K2.6.
- Kimi K2.6 is cheaper at $0.95 / $4 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.
- Kimi K2.6 has downloadable open weights; the other is API-only.

## Snapshot

| | GPT-5.3 Codex | Kimi K2.6 |
|---|---|---|
| Provider | OpenAI | Moonshot AI |
| Noometry Index | 45.5 | 47.5 |
| Rank | 61 | 52 |
| Context | 400K | 262K |
| Input $/M | $1.75 | $0.95 |
| Output $/M | $14 | $4 |
| Weights | Proprietary | Open |

## Coding

- GPT-5.3 Codex: 48.5 (#51)
- Kimi K2.6: 50.5 (#41)

| Benchmark | GPT-5.3 Codex | Kimi K2.6 |
|---|---|---|
| SWE-bench Verified | 74.8% | 76.7% |
| LMArena WebDev | 1408 | 1509 |
| WeirdML | 79.3% | 55.9% |
| ALE-Bench | 1,655 | 1,093 |
| SciCode | — | 53.5% |
| LMArena Coding | — | 1488 |

## Agentic & Tool Use

- GPT-5.3 Codex: 47.8 (#9)
- Kimi K2.6: 21.8 (#128)

| Benchmark | GPT-5.3 Codex | Kimi K2.6 |
|---|---|---|
| Vending-Bench 2 | 5,940 | 6,205 |
| Terminal-Bench | 78.4% | — |
| OSWorld 2.0 | — | 4.6% |
| ExploitBench | — | 18.4% |
| GBAEval | — | 0.9% |
| GDP.pdf | — | 12% |
| METR Time Horizons | 74.5% | — |

## Reasoning

- GPT-5.3 Codex: —
- Kimi K2.6: 39.6 (#51)

| Benchmark | GPT-5.3 Codex | Kimi K2.6 |
|---|---|---|
| Epoch Capabilities Index | 156.84 | 151.03 |
| NYT Connections (extended) | — | 87.2% |
| CritPt | — | 8% |
| Chess Puzzles | — | 26% |
| EBR-Bench | — | 2.4% |
| LMArena Hard Prompts | — | 1470 |
| Mystery Game Puzzles | — | 18% |
| DTBench | — | 90.9% |
| LMCA | — | 37.3% |

## Math

- GPT-5.3 Codex: —
- Kimi K2.6: 56.8 (#34)

| Benchmark | GPT-5.3 Codex | Kimi K2.6 |
|---|---|---|
| FrontierMath (Tiers 1-3) | — | 57.2% |
| FrontierMath Tier 4 | — | 25.6% |
| MathArena Final-Answer Competitions | — | 72.9% |
| OTIS Mock AIME 2024-2025 | — | 96.1% |
| ProofBench | — | 16% |
| LMArena Math | — | 1475 |
| FrontierMath (Feb 2025 set) | — | 39% |
| FrontierMath Tier 4 (v1) | — | 14.6% |

## Knowledge

- GPT-5.3 Codex: —
- Kimi K2.6: 53.9 (#48)

| Benchmark | GPT-5.3 Codex | Kimi K2.6 |
|---|---|---|
| GPQA Diamond | — | 90.8% |
| SimpleQA Verified | — | 34.9% |
| Vectara Hallucination Rate | — | 10.8% |
| LMArena Expert | — | 1491 |

## Multimodal

- GPT-5.3 Codex: —
- Kimi K2.6: 31.8 (#95)

| Benchmark | GPT-5.3 Codex | Kimi K2.6 |
|---|---|---|
| LMArena Vision | — | 1282 |
| Blueprint-Bench 2 | — | 3.9% |
| Furniture Assembly | — | 21.7% |
| LMArena Document | — | 1451 |

## Multilingual

- GPT-5.3 Codex: —
- Kimi K2.6: 55.1 (#35)

| Benchmark | GPT-5.3 Codex | Kimi K2.6 |
|---|---|---|
| LMArena Non-English | — | 1446 |
| LMArena Chinese | — | 1521 |
| LMArena French | — | 1471 |
| LMArena German | — | 1449 |
| LMArena Japanese | — | 1442 |
| LMArena Korean | — | 1428 |
| LMArena Russian | — | 1446 |
| LMArena Spanish | — | 1463 |

## Instruction Following

- GPT-5.3 Codex: —
- Kimi K2.6: 76.5 (#40)

| Benchmark | GPT-5.3 Codex | Kimi K2.6 |
|---|---|---|
| LMArena Instruction Following | — | 1451 |

## Long Context

- GPT-5.3 Codex: —
- Kimi K2.6: 44.9 (#50)

| Benchmark | GPT-5.3 Codex | Kimi K2.6 |
|---|---|---|
| LMArena Longer Query | — | 1468 |

## Writing & Preference

- GPT-5.3 Codex: —
- Kimi K2.6: 68.7 (#25)

| Benchmark | GPT-5.3 Codex | Kimi K2.6 |
|---|---|---|
| LMArena Text | — | 1455 |
| LMArena Creative Writing | — | 1434 |
| EQ-Bench Creative Writing | — | 1725 |
| EQ-Bench 4 | — | 1202 |
| LMArena Multi-Turn | — | 1453 |

## FAQ

### Is GPT-5.3 Codex better than Kimi K2.6?

Kimi K2.6 is the stronger model overall, scoring 47.5 to 45.5 on the Noometry Index.

### Which is cheaper, GPT-5.3 Codex or Kimi K2.6?

Kimi K2.6 is cheaper. It lists at $0.95 per million input tokens and $4 per million output tokens; GPT-5.3 Codex lists at $1.75 and $14.

### Is GPT-5.3 Codex or Kimi K2.6 better for coding?

Kimi K2.6 scores higher on coding benchmarks: 50.5 versus 48.5 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 Kimi K2.6 share?

6 benchmarks have published results for both models. GPT-5.3 Codex has 8 scored results on Noometry and Kimi K2.6 has 51.