# DeepSeek-V3.1-Terminus vs Kimi K3

> Kimi K3 is the stronger model overall, scoring 59.5 to 42.6 on the Noometry Index. DeepSeek-V3.1-Terminus costs 13× less per token, which makes it the better buy when Kimi K3's lead doesn't matter for your workload.

- Canonical page: https://noometry.com/compare/deepseek-v3-1-terminus-vs-kimi-k3
- Last updated: 2026-10-10
- Shared benchmarks: 15

## Summary

- They share 15 benchmarks with published results for both. DeepSeek-V3.1-Terminus scores higher in 0 categories and Kimi K3 in 7 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in math, where Kimi K3 leads 74.0 to 35.7.
- The biggest single-benchmark swing is LMCA: 28.6% for DeepSeek-V3.1-Terminus and 52.7% for Kimi K3.
- DeepSeek-V3.1-Terminus is cheaper at $0.30 / $1 per million input/output tokens, against $3 / $15 for Kimi K3.
- Kimi K3 accepts more context: 1.05M tokens versus 164K.

## Snapshot

| | DeepSeek-V3.1-Terminus | Kimi K3 |
|---|---|---|
| Provider | DeepSeek | Moonshot AI |
| Noometry Index | 42.6 | 59.5 |
| Rank | 96 | 12 |
| Context | 164K | 1.05M |
| Input $/M | $0.30 | $3 |
| Output $/M | $1 | $15 |
| Weights | Open | Open |

## Coding

- DeepSeek-V3.1-Terminus: 41.5 (#114)
- Kimi K3: 60.6 (#8)

| Benchmark | DeepSeek-V3.1-Terminus | Kimi K3 |
|---|---|---|
| SciCode | 40.6% | 59.5% |
| LMArena Coding | 1426 | 1509 |
| ALE-Bench | 745.17 | 1,524 |
| DeepSWE | — | 68.5% |
| FrontierCode | — | 44.2% |
| LMArena WebDev | — | 1659 |
| FrontierSWE | — | 25.9% |
| WeirdML | — | 82.6% |

## Agentic & Tool Use

- DeepSeek-V3.1-Terminus: —
- Kimi K3: 41.8 (#19)

| Benchmark | DeepSeek-V3.1-Terminus | Kimi K3 |
|---|---|---|
| APEX-Agents | — | 50.6% |
| τ²-bench Banking | — | 37.1% |
| PostTrainBench | — | 32% |
| GBAEval | — | 48.3% |
| GDP.pdf | — | 19% |
| Vending-Bench 2 | — | 5,165 |

## Reasoning

- DeepSeek-V3.1-Terminus: 25.7 (#126)
- Kimi K3: 63.0 (#15)

| Benchmark | DeepSeek-V3.1-Terminus | Kimi K3 |
|---|---|---|
| CritPt | 1.7% | 23.4% |
| LMArena Hard Prompts | 1427 | 1495 |
| DTBench | 81.3% | 91.2% |
| LMCA | 28.6% | 52.7% |
| ARC-AGI-2 | — | 60.4% |
| SimpleBench | — | 60.7% |
| Kagi LLM Benchmark | 57.4% | — |
| NYT Connections (extended) | — | 93.6% |
| ARC-AGI-1 | — | 94.5% |
| Chess Puzzles | — | 39% |
| Mystery Game Puzzles | — | 26% |
| Surface Evolver Bench | — | 95% |
| Epoch Capabilities Index | — | 157.68 |
| ForecastBench | — | 61.1 |

## Math

- DeepSeek-V3.1-Terminus: 35.7 (#138)
- Kimi K3: 74.0 (#13)

| Benchmark | DeepSeek-V3.1-Terminus | Kimi K3 |
|---|---|---|
| LMArena Math | 1402 | 1493 |
| FrontierMath (Tiers 1-3) | — | 72.2% |
| FrontierMath Tier 4 | — | 39% |
| MathArena Final-Answer Competitions | — | 87.8% |
| OTIS Mock AIME 2024-2025 | — | 97.2% |
| ProofBench | — | 87% |

## Knowledge

- DeepSeek-V3.1-Terminus: —
- Kimi K3: 63.2 (#19)

| Benchmark | DeepSeek-V3.1-Terminus | Kimi K3 |
|---|---|---|
| GPQA Diamond | — | 93.1% |
| SimpleQA Verified | — | 50.6% |
| LMArena Expert | — | 1520 |

## Multimodal

- DeepSeek-V3.1-Terminus: —
- Kimi K3: 38.4 (#58)

| Benchmark | DeepSeek-V3.1-Terminus | Kimi K3 |
|---|---|---|
| Blueprint-Bench 2 | — | 29.5% |
| Furniture Assembly | — | 34.2% |

## Multilingual

- DeepSeek-V3.1-Terminus: 52.3 (#87)
- Kimi K3: 56.5 (#20)

| Benchmark | DeepSeek-V3.1-Terminus | Kimi K3 |
|---|---|---|
| LMArena Non-English | 1407 | 1466 |
| LMArena Russian | 1434 | 1481 |
| LMArena Chinese | — | 1528 |
| LMArena French | — | 1488 |
| LMArena German | — | 1474 |
| LMArena Japanese | — | 1492 |
| LMArena Korean | — | 1465 |
| LMArena Spanish | — | 1478 |

## Instruction Following

- DeepSeek-V3.1-Terminus: 74.4 (#96)
- Kimi K3: 77.8 (#13)

| Benchmark | DeepSeek-V3.1-Terminus | Kimi K3 |
|---|---|---|
| LMArena Instruction Following | 1406 | 1481 |

## Long Context

- DeepSeek-V3.1-Terminus: 43.5 (#88)
- Kimi K3: 45.8 (#27)

| Benchmark | DeepSeek-V3.1-Terminus | Kimi K3 |
|---|---|---|
| LMArena Longer Query | 1426 | 1494 |

## Writing & Preference

- DeepSeek-V3.1-Terminus: 61.3 (#87)
- Kimi K3: 76.7 (#4)

| Benchmark | DeepSeek-V3.1-Terminus | Kimi K3 |
|---|---|---|
| LMArena Text | 1420 | 1476 |
| LMArena Creative Writing | 1404 | 1452 |
| LMArena Multi-Turn | 1411 | 1488 |
| EQ-Bench Creative Writing | — | 2082 |
| EQ-Bench 4 | — | 1339 |

## FAQ

### Is DeepSeek-V3.1-Terminus better than Kimi K3?

Kimi K3 is the stronger model overall, scoring 59.5 to 42.6 on the Noometry Index. DeepSeek-V3.1-Terminus costs 13× less per token, which makes it the better buy when Kimi K3's lead doesn't matter for your workload.

### Which is cheaper, DeepSeek-V3.1-Terminus or Kimi K3?

DeepSeek-V3.1-Terminus is cheaper. It lists at $0.30 per million input tokens and $1 per million output tokens; Kimi K3 lists at $3 and $15.

### Is DeepSeek-V3.1-Terminus or Kimi K3 better for coding?

Kimi K3 scores higher on coding benchmarks: 60.6 versus 41.5 in the Noometry coding category.

### Which has the bigger context window?

Kimi K3 does, with 1.05M tokens against 164K.

### How many benchmarks do DeepSeek-V3.1-Terminus and Kimi K3 share?

15 benchmarks have published results for both models. DeepSeek-V3.1-Terminus has 16 scored results on Noometry and Kimi K3 has 53.