# GPT-6.1 Sol vs Kimi K3

> GPT-6.1 Sol is the stronger model overall, scoring 65.6 to 59.5 on the Noometry Index.

- Canonical page: https://noometry.com/compare/gpt-6-1-sol-vs-kimi-k3
- Last updated: 2026-10-10
- Shared benchmarks: 32

## Summary

- They share 32 benchmarks with published results for both. GPT-6.1 Sol scores higher in 5 categories and Kimi K3 in 5 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-6.1 Sol leads 93.7 to 74.2.
- The biggest single-benchmark swing is FrontierMath Tier 4: 100% for GPT-6.1 Sol and 39% for Kimi K3.
- GPT-6.1 Sol is cheaper at $2 / $10 per million input/output tokens, against $3 / $15 for Kimi K3.
- GPT-6.1 Sol accepts more context: 1.05M tokens versus 1.05M.
- Kimi K3 has downloadable open weights; the other is API-only.

## Snapshot

| | GPT-6.1 Sol | Kimi K3 |
|---|---|---|
| Provider | OpenAI | Moonshot AI |
| Noometry Index | 65.6 | 59.5 |
| Rank | 6 | 15 |
| Context | 1.05M | 1.05M |
| Input $/M | $2 | $3 |
| Output $/M | $10 | $15 |
| Weights | Proprietary | Open |

## Coding

- GPT-6.1 Sol: 63.2 (#8)
- Kimi K3: 61.0 (#10)

| Benchmark | GPT-6.1 Sol | Kimi K3 |
|---|---|---|
| DeepSWE | 75.2% | 68.5% |
| FrontierCode | 50.2% | 44.2% |
| LMArena WebDev | 1755 | 1654 |
| SciCode | 55.8% | 59.5% |
| LMArena Coding | 1487 | 1508 |
| FrontierSWE | — | 25.9% |
| WeirdML | — | 82.6% |
| ALE-Bench | — | 1,524 |

## Agentic & Tool Use

- GPT-6.1 Sol: 39.6 (#26)
- Kimi K3: 41.8 (#20)

| Benchmark | GPT-6.1 Sol | Kimi K3 |
|---|---|---|
| APEX-Agents | 60% | 50.6% |
| GDP.pdf | 32% | 19% |
| τ²-bench Banking | — | 37.1% |
| PostTrainBench | — | 32% |
| GBAEval | — | 48.3% |
| Vending-Bench 2 | — | 5,165 |

## Reasoning

- GPT-6.1 Sol: 81.9 (#2)
- Kimi K3: 63.0 (#17)

| Benchmark | GPT-6.1 Sol | Kimi K3 |
|---|---|---|
| ARC-AGI-2 | 94.2% | 60.4% |
| NYT Connections (extended) | 95.5% | 93.6% |
| ARC-AGI-1 | 98.5% | 94.5% |
| CritPt | 31.7% | 23.4% |
| Chess Puzzles | 61% | 39% |
| LMArena Hard Prompts | 1466 | 1496 |
| Mystery Game Puzzles | 80% | 26% |
| Epoch Capabilities Index | 166.09 | 157.45 |
| SimpleBench | — | 60.7% |
| EBR-Bench | 54.3% | — |
| DTBench | — | 91.2% |
| LMCA | — | 52.7% |
| Surface Evolver Bench | — | 95% |
| ForecastBench | — | 61.1 |

## Math

- GPT-6.1 Sol: 93.7 (#1)
- Kimi K3: 74.2 (#16)

| Benchmark | GPT-6.1 Sol | Kimi K3 |
|---|---|---|
| FrontierMath (Tiers 1-3) | 93.7% | 72.2% |
| FrontierMath Tier 4 | 100% | 39% |
| OTIS Mock AIME 2024-2025 | 100% | 97.2% |
| ProofBench | 99% | 87% |
| LMArena Math | 1464 | 1491 |
| MathArena Final-Answer Competitions | — | 87.8% |

## Knowledge

- GPT-6.1 Sol: 71.8 (#4)
- Kimi K3: 63.2 (#21)

| Benchmark | GPT-6.1 Sol | Kimi K3 |
|---|---|---|
| GPQA Diamond | 95.4% | 93.1% |
| SimpleQA Verified | 73.9% | 50.6% |
| LMArena Expert | 1502 | 1521 |

## Multimodal

- GPT-6.1 Sol: 52.7 (#5)
- Kimi K3: 37.8 (#70)

| Benchmark | GPT-6.1 Sol | Kimi K3 |
|---|---|---|
| Furniture Assembly | 80% | 34.2% |
| LMArena Vision | 1288 | — |
| Blueprint-Bench 2 | — | 29.5% |

## Multilingual

- GPT-6.1 Sol: 54.3 (#46)
- Kimi K3: 56.3 (#21)

| Benchmark | GPT-6.1 Sol | Kimi K3 |
|---|---|---|
| LMArena Non-English | 1438 | 1466 |
| LMArena Chinese | 1477 | 1529 |
| LMArena Russian | 1455 | 1482 |
| LMArena French | — | 1491 |
| LMArena German | — | 1488 |
| LMArena Japanese | — | 1487 |
| LMArena Korean | — | 1458 |
| LMArena Spanish | — | 1472 |

## Instruction Following

- GPT-6.1 Sol: 77.0 (#29)
- Kimi K3: 77.7 (#14)

| Benchmark | GPT-6.1 Sol | Kimi K3 |
|---|---|---|
| LMArena Instruction Following | 1468 | 1483 |

## Long Context

- GPT-6.1 Sol: 44.9 (#54)
- Kimi K3: 45.8 (#29)

| Benchmark | GPT-6.1 Sol | Kimi K3 |
|---|---|---|
| LMArena Longer Query | 1465 | 1494 |

## Writing & Preference

- GPT-6.1 Sol: 63.6 (#63)
- Kimi K3: 76.6 (#4)

| Benchmark | GPT-6.1 Sol | Kimi K3 |
|---|---|---|
| LMArena Text | 1447 | 1476 |
| LMArena Creative Writing | 1432 | 1454 |
| LMArena Multi-Turn | 1449 | 1488 |
| EQ-Bench Creative Writing | — | 2082 |
| EQ-Bench 4 | — | 1339 |

## FAQ

### Is GPT-6.1 Sol better than Kimi K3?

GPT-6.1 Sol is the stronger model overall, scoring 65.6 to 59.5 on the Noometry Index.

### Which is cheaper, GPT-6.1 Sol or Kimi K3?

GPT-6.1 Sol is cheaper. It lists at $2 per million input tokens and $10 per million output tokens; Kimi K3 lists at $3 and $15.

### Is GPT-6.1 Sol or Kimi K3 better for coding?

GPT-6.1 Sol scores higher on coding benchmarks: 63.2 versus 61.0 in the Noometry coding category.

### Which has the bigger context window?

GPT-6.1 Sol does, with 1.05M tokens against 1.05M.

### How many benchmarks do GPT-6.1 Sol and Kimi K3 share?

32 benchmarks have published results for both models. GPT-6.1 Sol has 34 scored results on Noometry and Kimi K3 has 53.