# Granite 3.1 8b Instruct vs Inkling

> Inkling is the stronger model overall, scoring 44.1 to 32.4 on the Noometry Index.

- Canonical page: https://noometry.com/compare/granite-3-1-8b-instruct-vs-inkling
- Last updated: 2026-10-11
- Shared benchmarks: 12

## Summary

- They share 12 benchmarks with published results for both. Granite 3.1 8b Instruct scores higher in 2 categories and Inkling in 7 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Inkling leads 65.2 to 35.5.

## Snapshot

| | Granite 3.1 8b Instruct | Inkling |
|---|---|---|
| Provider | IBM | Thinking Machines Lab |
| Noometry Index | 32.4 | 44.1 |
| Rank | 258 | 80 |
| Context | — | 66K |
| Input $/M | — | $1.87 |
| Output $/M | — | $4.68 |
| Weights | Open | Open |

## Coding

- Granite 3.1 8b Instruct: 34.5 (#233)
- Inkling: 34.5 (#234)

| Benchmark | Granite 3.1 8b Instruct | Inkling |
|---|---|---|
| LMArena Coding | 1186 | 1464 |
| FrontierCode | — | 14% |
| LMArena WebDev | — | 1413 |
| FrontierSWE | — | 4.1% |
| SciCode | — | 47% |
| WeirdML | — | 32.3% |
| ALE-Bench | — | 946 |

## Agentic & Tool Use

- Granite 3.1 8b Instruct: 24.1 (#120)
- Inkling: 29.6 (#85)

| Benchmark | Granite 3.1 8b Instruct | Inkling |
|---|---|---|
| APEX-Agents | — | 33.8% |
| Berkeley Function Calling Leaderboard | 27.1% | — |
| τ²-bench Banking | — | 25% |

## Reasoning

- Granite 3.1 8b Instruct: 22.1 (#207)
- Inkling: 40.4 (#56)

| Benchmark | Granite 3.1 8b Instruct | Inkling |
|---|---|---|
| LMArena Hard Prompts | 1145 | 1451 |
| ARC-AGI-2 | — | 36.5% |
| SimpleBench | — | 50% |
| ARC-AGI-1 | — | 79.5% |
| CritPt | — | 5.4% |
| Chess Puzzles | — | 21% |
| DTBench | — | 87.5% |
| LMCA | — | 37.6% |
| Epoch Capabilities Index | — | 148.54 |

## Math

- Granite 3.1 8b Instruct: 33.0 (#209)
- Inkling: 31.3 (#225)

| Benchmark | Granite 3.1 8b Instruct | Inkling |
|---|---|---|
| LMArena Math | 1152 | 1479 |
| FrontierMath (Tiers 1-3) | — | 33.3% |
| FrontierMath Tier 4 | — | 4.9% |
| OTIS Mock AIME 2024-2025 | — | 88.9% |
| ProofBench | — | 0% |

## Knowledge

- Granite 3.1 8b Instruct: 31.1 (#220)
- Inkling: 55.1 (#49)

| Benchmark | Granite 3.1 8b Instruct | Inkling |
|---|---|---|
| LMArena Expert | 1142 | 1465 |
| GPQA Diamond | — | 88.3% |
| SimpleQA Verified | — | 40.3% |

## Multilingual

- Granite 3.1 8b Instruct: 30.9 (#260)
- Inkling: 54.0 (#52)

| Benchmark | Granite 3.1 8b Instruct | Inkling |
|---|---|---|
| LMArena Non-English | 1099 | 1434 |
| LMArena Chinese | 1145 | 1490 |
| LMArena Russian | 1092 | 1429 |
| LMArena French | — | 1458 |
| LMArena German | — | 1446 |
| LMArena Japanese | — | 1429 |
| LMArena Korean | — | 1404 |
| LMArena Spanish | — | 1448 |

## Instruction Following

- Granite 3.1 8b Instruct: 58.6 (#259)
- Inkling: 75.1 (#71)

| Benchmark | Granite 3.1 8b Instruct | Inkling |
|---|---|---|
| LMArena Instruction Following | 1131 | 1426 |

## Long Context

- Granite 3.1 8b Instruct: 35.2 (#241)
- Inkling: 43.8 (#86)

| Benchmark | Granite 3.1 8b Instruct | Inkling |
|---|---|---|
| LMArena Longer Query | 1162 | 1434 |

## Writing & Preference

- Granite 3.1 8b Instruct: 35.5 (#266)
- Inkling: 65.2 (#51)

| Benchmark | Granite 3.1 8b Instruct | Inkling |
|---|---|---|
| LMArena Text | 1150 | 1441 |
| LMArena Creative Writing | 1129 | 1387 |
| LMArena Multi-Turn | 1108 | 1436 |
| EQ-Bench Creative Writing | — | 1611 |
| EQ-Bench 4 | — | 1226 |

## FAQ

### Is Granite 3.1 8b Instruct better than Inkling?

Inkling is the stronger model overall, scoring 44.1 to 32.4 on the Noometry Index.

### Is Granite 3.1 8b Instruct or Inkling better for coding?

They score almost the same on coding (34.5 vs 34.5); test both on your own repository before choosing.

### How many benchmarks do Granite 3.1 8b Instruct and Inkling share?

12 benchmarks have published results for both models. Granite 3.1 8b Instruct has 13 scored results on Noometry and Inkling has 41.