# Inkling-Small vs Llama 3-70B

> Inkling-Small is the stronger model overall, scoring 46.5 to 28.5 on the Noometry Index.

- Canonical page: https://noometry.com/compare/inkling-small-vs-llama-3-70b
- Last updated: 2026-10-10
- Shared benchmarks: 20

## Summary

- They share 20 benchmarks with published results for both. Inkling-Small scores higher in 8 categories and Llama 3-70B in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Inkling-Small leads 44.9 to 12.0.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 90% for Inkling-Small and 4.3% for Llama 3-70B.

## Snapshot

| | Inkling-Small | Llama 3-70B |
|---|---|---|
| Provider | Thinking Machines Lab | Meta |
| Noometry Index | 46.5 | 28.5 |
| Rank | 57 | 317 |
| Context | 524K | — |
| Input $/M | $0.45 | — |
| Output $/M | $1.20 | — |
| Weights | Open | Open |

## Coding

- Inkling-Small: 43.3 (#80)
- Llama 3-70B: 35.3 (#215)

| Benchmark | Inkling-Small | Llama 3-70B |
|---|---|---|
| LMArena Coding | 1451 | 1206 |
| LMArena WebDev | 1408 | — |
| SciCode | 48.7% | — |
| BigCodeBench Instruct | — | 43.6% |
| BigCodeBench Complete | — | 54.5% |
| HumanEval+ | — | 72% |
| MBPP+ | — | 69% |

## Agentic & Tool Use

- Inkling-Small: —
- Llama 3-70B: 20.9 (#132)

| Benchmark | Inkling-Small | Llama 3-70B |
|---|---|---|
| Cybench | — | 5% |

## Reasoning

- Inkling-Small: 39.0 (#53)
- Llama 3-70B: 17.4 (#284)

| Benchmark | Inkling-Small | Llama 3-70B |
|---|---|---|
| LMArena Hard Prompts | 1424 | 1195 |
| Epoch Capabilities Index | 150.13 | 122.91 |
| ARC-AGI-2 | 40.1% | — |
| Kagi LLM Benchmark | — | 35.1% |
| ARC-AGI-1 | 84% | — |
| CritPt | 8.3% | — |
| Chess Puzzles | 18% | — |
| Mystery Game Puzzles | 6% | — |
| DTBench | — | 54.2% |
| ForecastBench | — | 57.1 |
| WinoGrande | — | 83.5% |

## Math

- Inkling-Small: 44.9 (#65)
- Llama 3-70B: 12.0 (#299)

| Benchmark | Inkling-Small | Llama 3-70B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 90% | 4.3% |
| LMArena Math | 1453 | 1218 |
| FrontierMath (Tiers 1-3) | 46.3% | — |
| FrontierMath Tier 4 | 17.1% | — |
| ProofBench | 6% | — |
| MATH Level 5 | — | 22.6% |

## Knowledge

- Inkling-Small: 48.3 (#71)
- Llama 3-70B: 20.8 (#272)

| Benchmark | Inkling-Small | Llama 3-70B |
|---|---|---|
| GPQA Diamond | 88.5% | 40.6% |
| LMArena Expert | 1442 | 1149 |
| SimpleQA Verified | 19.1% | — |
| MMLU | — | 79.3% |

## Multimodal

- Inkling-Small: 38.4 (#59)
- Llama 3-70B: —

| Benchmark | Inkling-Small | Llama 3-70B |
|---|---|---|
| LMArena Vision | 1235 | — |

## Multilingual

- Inkling-Small: 51.7 (#102)
- Llama 3-70B: 33.8 (#246)

| Benchmark | Inkling-Small | Llama 3-70B |
|---|---|---|
| LMArena Non-English | 1400 | 1142 |
| LMArena Chinese | 1458 | 1114 |
| LMArena French | 1438 | 1231 |
| LMArena German | 1412 | 1169 |
| LMArena Japanese | 1399 | 1017 |
| LMArena Korean | 1355 | 1017 |
| LMArena Russian | 1392 | 1159 |
| LMArena Spanish | 1426 | 1241 |

## Instruction Following

- Inkling-Small: 73.9 (#112)
- Llama 3-70B: 62.8 (#233)

| Benchmark | Inkling-Small | Llama 3-70B |
|---|---|---|
| LMArena Instruction Following | 1397 | 1194 |

## Long Context

- Inkling-Small: 42.7 (#113)
- Llama 3-70B: 35.6 (#235)

| Benchmark | Inkling-Small | Llama 3-70B |
|---|---|---|
| LMArena Longer Query | 1401 | 1174 |

## Writing & Preference

- Inkling-Small: 59.8 (#102)
- Llama 3-70B: 43.0 (#225)

| Benchmark | Inkling-Small | Llama 3-70B |
|---|---|---|
| LMArena Text | 1413 | 1221 |
| LMArena Creative Writing | 1330 | 1210 |
| LMArena Multi-Turn | 1420 | 1223 |
| EQ-Bench Creative Writing | 1491 | — |

## FAQ

### Is Inkling-Small better than Llama 3-70B?

Inkling-Small is the stronger model overall, scoring 46.5 to 28.5 on the Noometry Index.

### Is Inkling-Small or Llama 3-70B better for coding?

Inkling-Small scores higher on coding benchmarks: 43.3 versus 35.3 in the Noometry coding category.

### How many benchmarks do Inkling-Small and Llama 3-70B share?

20 benchmarks have published results for both models. Inkling-Small has 33 scored results on Noometry and Llama 3-70B has 31.