# GPT-4.5 vs Kimi K2.5

> Kimi K2.5 is the stronger model overall, scoring 47.8 to 36.9 on the Noometry Index.

- Canonical page: https://noometry.com/compare/gpt-4-5-vs-kimi-k2-5
- Last updated: 2026-10-10
- Shared benchmarks: 28

## Summary

- They share 28 benchmarks with published results for both. GPT-4.5 scores higher in 0 categories and Kimi K2.5 in 10 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Kimi K2.5 leads 53.6 to 32.4.
- The biggest single-benchmark swing is ARC-AGI-1: 10.3% for GPT-4.5 and 65.3% for Kimi K2.5.
- Kimi K2.5 has downloadable open weights; the other is API-only.

## Snapshot

| | GPT-4.5 | Kimi K2.5 |
|---|---|---|
| Provider | OpenAI | Moonshot AI |
| Noometry Index | 36.9 | 47.8 |
| Rank | 200 | 50 |
| Context | — | 262K |
| Input $/M | — | $0.45 |
| Output $/M | — | $2.25 |
| Weights | Proprietary | Open |

## Coding

- GPT-4.5: 41.8 (#102)
- Kimi K2.5: 48.7 (#48)

| Benchmark | GPT-4.5 | Kimi K2.5 |
|---|---|---|
| WeirdML | 39.4% | 45.6% |
| LMArena Coding | 1397 | 1474 |
| SWE-bench Verified | — | 73.8% |
| SWE-bench Verified (bash only) | — | 70.8% |
| Aider Polyglot | 44.9% | — |
| LMArena WebDev | — | 1436 |
| SWE-bench Multilingual | — | 67.3% |
| SciCode | — | 49% |
| LiveBench Coding | 75.2% | — |
| ALE-Bench | — | 821.65 |

## Agentic & Tool Use

- GPT-4.5: 27.6 (#92)
- Kimi K2.5: 34.0 (#44)

| Benchmark | GPT-4.5 | Kimi K2.5 |
|---|---|---|
| Terminal-Bench | — | 43.2% |
| Cybench | 17.5% | — |
| OSWorld | — | 63.3% |
| Vending-Bench 2 | — | 1,198 |

## Reasoning

- GPT-4.5: 14.1 (#320)
- Kimi K2.5: 31.2 (#74)

| Benchmark | GPT-4.5 | Kimi K2.5 |
|---|---|---|
| ARC-AGI-2 | 0.8% | 11.8% |
| SimpleBench | 34.5% | 46.8% |
| ARC-AGI-1 | 10.3% | 65.3% |
| EnigmaEval | 3.2% | 3.4% |
| LMArena Hard Prompts | 1404 | 1454 |
| Epoch Capabilities Index | 136.74 | 148.05 |
| Kagi LLM Benchmark | — | 78.5% |
| NYT Connections (extended) | — | 69.9% |
| CritPt | — | 3.1% |
| Chess Puzzles | — | 12% |
| Thematic Generalization | — | 69.4% |
| LiveBench Reasoning | 71.1% | — |
| LiveBench Data Analysis | 64.3% | — |
| ForecastBench | 61.7 | — |
| LiveBench | 69% | — |

## Math

- GPT-4.5: 31.0 (#200)
- Kimi K2.5: 50.3 (#46)

| Benchmark | GPT-4.5 | Kimi K2.5 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 37.8% | 92.2% |
| LMArena Math | 1412 | 1470 |
| MathArena Final-Answer Competitions | — | 62.3% |
| LiveBench Math | 69.3% | — |
| MATH Level 5 | 78.6% | — |
| FrontierMath (Feb 2025 set) | — | 27.9% |
| FrontierMath Tier 4 (v1) | — | 4.2% |

## Knowledge

- GPT-4.5: 32.4 (#205)
- Kimi K2.5: 53.6 (#52)

| Benchmark | GPT-4.5 | Kimi K2.5 |
|---|---|---|
| GPQA Diamond | 68.7% | 87.6% |
| Humanity's Last Exam | 5.4% | 24.4% |
| LMArena Expert | 1394 | 1466 |
| SimpleQA Verified | — | 34.3% |
| Confabulations | 13.6% | — |
| Vectara Hallucination Rate | — | 14.2% |

## Multimodal

- GPT-4.5: 37.0 (#68)
- Kimi K2.5: 40.4 (#37)

| Benchmark | GPT-4.5 | Kimi K2.5 |
|---|---|---|
| LMArena Vision | 1195 | 1269 |
| VPCT | 45% | — |
| LMArena Document | — | 1430 |

## Multilingual

- GPT-4.5: 52.7 (#78)
- Kimi K2.5: 54.1 (#50)

| Benchmark | GPT-4.5 | Kimi K2.5 |
|---|---|---|
| LMArena Non-English | 1413 | 1433 |
| LMArena Chinese | 1421 | 1495 |
| LMArena French | 1417 | 1454 |
| LMArena German | 1457 | 1441 |
| LMArena Japanese | 1416 | 1422 |
| LMArena Korean | 1392 | 1412 |
| LMArena Russian | 1419 | 1436 |
| LMArena Spanish | — | 1450 |

## Instruction Following

- GPT-4.5: 72.8 (#129)
- Kimi K2.5: 75.5 (#60)

| Benchmark | GPT-4.5 | Kimi K2.5 |
|---|---|---|
| LMArena Instruction Following | 1404 | 1431 |
| LiveBench Instruction Following | 72.3% | — |

## Long Context

- GPT-4.5: 40.4 (#150)
- Kimi K2.5: 52.1 (#7)

| Benchmark | GPT-4.5 | Kimi K2.5 |
|---|---|---|
| Fiction.LiveBench | 63.9% | 86.1% |
| LMArena Longer Query | 1406 | 1445 |
| CL-bench | — | 19.3% |
| CL-bench Life | — | 13.2% |

## Writing & Preference

- GPT-4.5: 57.1 (#129)
- Kimi K2.5: 65.3 (#51)

| Benchmark | GPT-4.5 | Kimi K2.5 |
|---|---|---|
| LMArena Text | 1417 | 1445 |
| LMArena Creative Writing | 1394 | 1423 |
| EQ-Bench Creative Writing | 1258 | 1579 |
| LMArena Multi-Turn | 1444 | 1444 |
| Short-Story Creative Writing | 75.6% | — |
| LiveBench Language | 61.5% | — |

## FAQ

### Is GPT-4.5 better than Kimi K2.5?

Kimi K2.5 is the stronger model overall, scoring 47.8 to 36.9 on the Noometry Index.

### Is GPT-4.5 or Kimi K2.5 better for coding?

Kimi K2.5 scores higher on coding benchmarks: 48.7 versus 41.8 in the Noometry coding category.

### How many benchmarks do GPT-4.5 and Kimi K2.5 share?

28 benchmarks have published results for both models. GPT-4.5 has 42 scored results on Noometry and Kimi K2.5 has 51.