# Granite 3.1 8b Instruct vs Step 3

> Step 3 is the stronger model overall, scoring 40.5 to 32.4 on the Noometry Index.

- Canonical page: https://noometry.com/compare/granite-3-1-8b-instruct-vs-step-3
- 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 0 categories and Step 3 in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Step 3 leads 54.3 to 35.5.

## Snapshot

| | Granite 3.1 8b Instruct | Step 3 |
|---|---|---|
| Provider | IBM | StepFun |
| Noometry Index | 32.4 | 40.5 |
| Rank | 258 | 149 |
| Context | — | — |
| Input $/M | — | — |
| Output $/M | — | — |
| Weights | Open | Open |

## Coding

- Granite 3.1 8b Instruct: 34.5 (#233)
- Step 3: 40.1 (#147)

| Benchmark | Granite 3.1 8b Instruct | Step 3 |
|---|---|---|
| LMArena Coding | 1186 | 1367 |

## Agentic & Tool Use

- Granite 3.1 8b Instruct: 24.1 (#120)
- Step 3: —

| Benchmark | Granite 3.1 8b Instruct | Step 3 |
|---|---|---|
| Berkeley Function Calling Leaderboard | 27.1% | — |

## Reasoning

- Granite 3.1 8b Instruct: 22.1 (#207)
- Step 3: 28.4 (#105)

| Benchmark | Granite 3.1 8b Instruct | Step 3 |
|---|---|---|
| LMArena Hard Prompts | 1145 | 1355 |
| Kagi LLM Benchmark | — | 62.3% |

## Math

- Granite 3.1 8b Instruct: 33.0 (#209)
- Step 3: 37.6 (#148)

| Benchmark | Granite 3.1 8b Instruct | Step 3 |
|---|---|---|
| LMArena Math | 1152 | 1366 |

## Knowledge

- Granite 3.1 8b Instruct: 31.1 (#220)
- Step 3: 36.8 (#164)

| Benchmark | Granite 3.1 8b Instruct | Step 3 |
|---|---|---|
| LMArena Expert | 1142 | 1333 |

## Multimodal

- Granite 3.1 8b Instruct: —
- Step 3: 35.5 (#86)

| Benchmark | Granite 3.1 8b Instruct | Step 3 |
|---|---|---|
| LMArena Vision | — | 1177 |

## Multilingual

- Granite 3.1 8b Instruct: 30.9 (#260)
- Step 3: 46.3 (#159)

| Benchmark | Granite 3.1 8b Instruct | Step 3 |
|---|---|---|
| LMArena Non-English | 1099 | 1327 |
| LMArena Chinese | 1145 | 1397 |
| LMArena Russian | 1092 | 1331 |
| LMArena German | — | 1371 |
| LMArena Korean | — | 1269 |
| LMArena Spanish | — | 1371 |

## Instruction Following

- Granite 3.1 8b Instruct: 58.6 (#259)
- Step 3: 70.4 (#164)

| Benchmark | Granite 3.1 8b Instruct | Step 3 |
|---|---|---|
| LMArena Instruction Following | 1131 | 1332 |

## Long Context

- Granite 3.1 8b Instruct: 35.2 (#241)
- Step 3: 40.3 (#157)

| Benchmark | Granite 3.1 8b Instruct | Step 3 |
|---|---|---|
| LMArena Longer Query | 1162 | 1326 |

## Writing & Preference

- Granite 3.1 8b Instruct: 35.5 (#266)
- Step 3: 54.3 (#151)

| Benchmark | Granite 3.1 8b Instruct | Step 3 |
|---|---|---|
| LMArena Text | 1150 | 1350 |
| LMArena Creative Writing | 1129 | 1321 |
| LMArena Multi-Turn | 1108 | 1341 |

## FAQ

### Is Granite 3.1 8b Instruct better than Step 3?

Step 3 is the stronger model overall, scoring 40.5 to 32.4 on the Noometry Index.

### Is Granite 3.1 8b Instruct or Step 3 better for coding?

Step 3 scores higher on coding benchmarks: 40.1 versus 34.5 in the Noometry coding category.

### How many benchmarks do Granite 3.1 8b Instruct and Step 3 share?

12 benchmarks have published results for both models. Granite 3.1 8b Instruct has 13 scored results on Noometry and Step 3 has 17.