# Kimi K2.5 vs Llama 3-70B

> Kimi K2.5 is the stronger model overall, scoring 47.8 to 28.5 on the Noometry Index.

- Canonical page: https://noometry.com/compare/kimi-k2-5-vs-llama-3-70b
- Last updated: 2026-10-10
- Shared benchmarks: 21

## Summary

- They share 21 benchmarks with published results for both. Kimi K2.5 scores higher in 9 categories and Llama 3-70B in 0 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where Kimi K2.5 leads 50.3 to 12.0.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 92.2% for Kimi K2.5 and 4.3% for Llama 3-70B.

## Snapshot

| | Kimi K2.5 | Llama 3-70B |
|---|---|---|
| Provider | Moonshot AI | Meta |
| Noometry Index | 47.8 | 28.5 |
| Rank | 50 | 317 |
| Context | 262K | — |
| Input $/M | $0.45 | — |
| Output $/M | $2.25 | — |
| Weights | Open | Open |

## Coding

- Kimi K2.5: 48.7 (#48)
- Llama 3-70B: 35.3 (#215)

| Benchmark | Kimi K2.5 | Llama 3-70B |
|---|---|---|
| LMArena Coding | 1474 | 1206 |
| SWE-bench Verified | 73.8% | — |
| SWE-bench Verified (bash only) | 70.8% | — |
| LMArena WebDev | 1436 | — |
| SWE-bench Multilingual | 67.3% | — |
| SciCode | 49% | — |
| WeirdML | 45.6% | — |
| BigCodeBench Instruct | — | 43.6% |
| BigCodeBench Complete | — | 54.5% |
| ALE-Bench | 821.65 | — |
| HumanEval+ | — | 72% |
| MBPP+ | — | 69% |

## Agentic & Tool Use

- Kimi K2.5: 34.0 (#44)
- Llama 3-70B: 20.9 (#132)

| Benchmark | Kimi K2.5 | Llama 3-70B |
|---|---|---|
| Terminal-Bench | 43.2% | — |
| Cybench | — | 5% |
| OSWorld | 63.3% | — |
| Vending-Bench 2 | 1,198 | — |

## Reasoning

- Kimi K2.5: 31.2 (#74)
- Llama 3-70B: 17.4 (#284)

| Benchmark | Kimi K2.5 | Llama 3-70B |
|---|---|---|
| Kagi LLM Benchmark | 78.5% | 35.1% |
| LMArena Hard Prompts | 1454 | 1195 |
| Epoch Capabilities Index | 148.05 | 122.91 |
| ARC-AGI-2 | 11.8% | — |
| SimpleBench | 46.8% | — |
| NYT Connections (extended) | 69.9% | — |
| ARC-AGI-1 | 65.3% | — |
| CritPt | 3.1% | — |
| Chess Puzzles | 12% | — |
| EnigmaEval | 3.4% | — |
| Thematic Generalization | 69.4% | — |
| DTBench | — | 54.2% |
| ForecastBench | — | 57.1 |
| WinoGrande | — | 83.5% |

## Math

- Kimi K2.5: 50.3 (#46)
- Llama 3-70B: 12.0 (#299)

| Benchmark | Kimi K2.5 | Llama 3-70B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 92.2% | 4.3% |
| LMArena Math | 1470 | 1218 |
| MathArena Final-Answer Competitions | 62.3% | — |
| MATH Level 5 | — | 22.6% |
| FrontierMath (Feb 2025 set) | 27.9% | — |
| FrontierMath Tier 4 (v1) | 4.2% | — |

## Knowledge

- Kimi K2.5: 53.6 (#52)
- Llama 3-70B: 20.8 (#272)

| Benchmark | Kimi K2.5 | Llama 3-70B |
|---|---|---|
| GPQA Diamond | 87.6% | 40.6% |
| LMArena Expert | 1466 | 1149 |
| Humanity's Last Exam | 24.4% | — |
| SimpleQA Verified | 34.3% | — |
| Vectara Hallucination Rate | 14.2% | — |
| MMLU | — | 79.3% |

## Multimodal

- Kimi K2.5: 40.4 (#37)
- Llama 3-70B: —

| Benchmark | Kimi K2.5 | Llama 3-70B |
|---|---|---|
| LMArena Vision | 1269 | — |
| LMArena Document | 1430 | — |

## Multilingual

- Kimi K2.5: 54.1 (#50)
- Llama 3-70B: 33.8 (#246)

| Benchmark | Kimi K2.5 | Llama 3-70B |
|---|---|---|
| LMArena Non-English | 1433 | 1142 |
| LMArena Chinese | 1495 | 1114 |
| LMArena French | 1454 | 1231 |
| LMArena German | 1441 | 1169 |
| LMArena Japanese | 1422 | 1017 |
| LMArena Korean | 1412 | 1017 |
| LMArena Russian | 1436 | 1159 |
| LMArena Spanish | 1450 | 1241 |

## Instruction Following

- Kimi K2.5: 75.5 (#60)
- Llama 3-70B: 62.8 (#233)

| Benchmark | Kimi K2.5 | Llama 3-70B |
|---|---|---|
| LMArena Instruction Following | 1431 | 1194 |

## Long Context

- Kimi K2.5: 52.1 (#7)
- Llama 3-70B: 35.6 (#235)

| Benchmark | Kimi K2.5 | Llama 3-70B |
|---|---|---|
| LMArena Longer Query | 1445 | 1174 |
| Fiction.LiveBench | 86.1% | — |
| CL-bench | 19.3% | — |
| CL-bench Life | 13.2% | — |

## Writing & Preference

- Kimi K2.5: 65.3 (#51)
- Llama 3-70B: 43.0 (#225)

| Benchmark | Kimi K2.5 | Llama 3-70B |
|---|---|---|
| LMArena Text | 1445 | 1221 |
| LMArena Creative Writing | 1423 | 1210 |
| LMArena Multi-Turn | 1444 | 1223 |
| EQ-Bench Creative Writing | 1579 | — |

## FAQ

### Is Kimi K2.5 better than Llama 3-70B?

Kimi K2.5 is the stronger model overall, scoring 47.8 to 28.5 on the Noometry Index.

### Is Kimi K2.5 or Llama 3-70B better for coding?

Kimi K2.5 scores higher on coding benchmarks: 48.7 versus 35.3 in the Noometry coding category.

### How many benchmarks do Kimi K2.5 and Llama 3-70B share?

21 benchmarks have published results for both models. Kimi K2.5 has 51 scored results on Noometry and Llama 3-70B has 31.