# Inkling vs Llama 2-70B

> Inkling is the stronger model overall, scoring 44.1 to 24.4 on the Noometry Index.

- Canonical page: https://noometry.com/compare/inkling-vs-llama-2-70b
- Last updated: 2026-10-10
- Shared benchmarks: 21

## Summary

- They share 21 benchmarks with published results for both. Inkling scores higher in 8 categories and Llama 2-70B in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Inkling leads 55.1 to 7.4.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 88.9% for Inkling and 0% for Llama 2-70B.

## Snapshot

| | Inkling | Llama 2-70B |
|---|---|---|
| Provider | Thinking Machines Lab | Meta |
| Noometry Index | 44.1 | 24.4 |
| Rank | 80 | 349 |
| Context | 66K | — |
| Input $/M | $1.87 | — |
| Output $/M | $4.68 | — |
| Weights | Open | Open |

## Coding

- Inkling: 34.5 (#234)
- Llama 2-70B: 31.4 (#286)

| Benchmark | Inkling | Llama 2-70B |
|---|---|---|
| LMArena Coding | 1464 | 1079 |
| FrontierCode | 14% | — |
| LMArena WebDev | 1413 | — |
| FrontierSWE | 4.1% | — |
| SciCode | 47% | — |
| WeirdML | 32.3% | — |
| ALE-Bench | 946 | — |

## Agentic & Tool Use

- Inkling: 29.6 (#85)
- Llama 2-70B: —

| Benchmark | Inkling | Llama 2-70B |
|---|---|---|
| APEX-Agents | 33.8% | — |
| τ²-bench Banking | 25% | — |

## Reasoning

- Inkling: 40.4 (#56)
- Llama 2-70B: 14.4 (#325)

| Benchmark | Inkling | Llama 2-70B |
|---|---|---|
| LMArena Hard Prompts | 1451 | 1073 |
| DTBench | 87.5% | 41.6% |
| Epoch Capabilities Index | 148.54 | 113.79 |
| ARC-AGI-2 | 36.5% | — |
| SimpleBench | 50% | — |
| ARC-AGI-1 | 79.5% | — |
| CritPt | 5.4% | — |
| Chess Puzzles | 21% | — |
| LMCA | 37.6% | — |
| BIG-Bench Hard | — | 64.9% |
| CommonsenseQA 2.0 | — | 50% |
| ForecastBench | — | 51.4 |
| HellaSwag | — | 85.3% |
| LAMBADA | — | 78.9% |
| PIQA | — | 82.8% |
| WinoGrande | — | 80.2% |

## Math

- Inkling: 31.3 (#225)
- Llama 2-70B: 8.1 (#326)

| Benchmark | Inkling | Llama 2-70B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 88.9% | 0% |
| LMArena Math | 1479 | 1091 |
| FrontierMath (Tiers 1-3) | 33.3% | — |
| FrontierMath Tier 4 | 4.9% | — |
| ProofBench | 0% | — |
| MATH Level 5 | — | 3.3% |
| GSM8K | — | 69.6% |

## Knowledge

- Inkling: 55.1 (#49)
- Llama 2-70B: 7.4 (#310)

| Benchmark | Inkling | Llama 2-70B |
|---|---|---|
| GPQA Diamond | 88.3% | 26.3% |
| LMArena Expert | 1465 | 1039 |
| SimpleQA Verified | 40.3% | — |
| ARC (AI2) Challenge | — | 78.3% |
| BoolQ | — | 88.6% |
| MMLU | — | 69.9% |
| OpenBookQA | — | 60.2% |
| TriviaQA | — | 87.6% |

## Multilingual

- Inkling: 54.0 (#52)
- Llama 2-70B: 27.7 (#274)

| Benchmark | Inkling | Llama 2-70B |
|---|---|---|
| LMArena Non-English | 1434 | 1045 |
| LMArena Chinese | 1490 | 995 |
| LMArena French | 1458 | 1090 |
| LMArena German | 1446 | 1041 |
| LMArena Japanese | 1429 | 927 |
| LMArena Korean | 1404 | 964 |
| LMArena Russian | 1429 | 1083 |
| LMArena Spanish | 1448 | 1143 |

## Instruction Following

- Inkling: 75.1 (#71)
- Llama 2-70B: 54.9 (#278)

| Benchmark | Inkling | Llama 2-70B |
|---|---|---|
| LMArena Instruction Following | 1426 | 1071 |

## Long Context

- Inkling: 43.8 (#86)
- Llama 2-70B: 32.3 (#270)

| Benchmark | Inkling | Llama 2-70B |
|---|---|---|
| LMArena Longer Query | 1434 | 1062 |

## Writing & Preference

- Inkling: 65.2 (#51)
- Llama 2-70B: 32.3 (#279)

| Benchmark | Inkling | Llama 2-70B |
|---|---|---|
| LMArena Text | 1441 | 1115 |
| LMArena Creative Writing | 1387 | 1075 |
| LMArena Multi-Turn | 1436 | 1088 |
| EQ-Bench Creative Writing | 1611 | — |
| EQ-Bench 4 | 1226 | — |

## FAQ

### Is Inkling better than Llama 2-70B?

Inkling is the stronger model overall, scoring 44.1 to 24.4 on the Noometry Index.

### Is Inkling or Llama 2-70B better for coding?

Inkling scores higher on coding benchmarks: 34.5 versus 31.4 in the Noometry coding category.

### How many benchmarks do Inkling and Llama 2-70B share?

21 benchmarks have published results for both models. Inkling has 41 scored results on Noometry and Llama 2-70B has 35.