# DeepSeek-V2.5 (Sep 2024) vs Granite 3.1 8b Instruct

> DeepSeek-V2.5 (Sep 2024) is the stronger model overall, scoring 37.0 to 31.8 on the Noometry Index.

- Canonical page: https://noometry.com/compare/deepseek-v2-5-vs-granite-3-1-8b-instruct
- Last updated: 2026-10-10
- Shared benchmarks: 12

## Summary

- They share 12 benchmarks with published results for both. DeepSeek-V2.5 (Sep 2024) scores higher in 7 categories and Granite 3.1 8b Instruct in 1 category; 8 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where DeepSeek-V2.5 (Sep 2024) leads 50.0 to 35.7.

## Snapshot

| | DeepSeek-V2.5 (Sep 2024) | Granite 3.1 8b Instruct |
|---|---|---|
| Provider | DeepSeek | IBM |
| Noometry Index | 37.0 | 31.8 |
| Rank | 197 | 256 |
| Context | — | — |
| Input $/M | — | — |
| Output $/M | — | — |
| Weights | Open | Open |

## Coding

- DeepSeek-V2.5 (Sep 2024): 31.3 (#274)
- Granite 3.1 8b Instruct: 33.9 (#232)

| Benchmark | DeepSeek-V2.5 (Sep 2024) | Granite 3.1 8b Instruct |
|---|---|---|
| LMArena Coding | 1310 | 1186 |
| Aider Polyglot | 17.8% | — |
| BigCodeBench Instruct | 48.6% | — |
| BigCodeBench Complete | 53.2% | — |
| HumanEval+ | 83.5% | — |
| MBPP+ | 74.1% | — |

## Agentic & Tool Use

- DeepSeek-V2.5 (Sep 2024): —
- Granite 3.1 8b Instruct: 23.9 (#113)

| Benchmark | DeepSeek-V2.5 (Sep 2024) | Granite 3.1 8b Instruct |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 27.1% |

## Reasoning

- DeepSeek-V2.5 (Sep 2024): 24.5 (#145)
- Granite 3.1 8b Instruct: 21.2 (#204)

| Benchmark | DeepSeek-V2.5 (Sep 2024) | Granite 3.1 8b Instruct |
|---|---|---|
| LMArena Hard Prompts | 1289 | 1146 |

## Math

- DeepSeek-V2.5 (Sep 2024): 33.3 (#177)
- Granite 3.1 8b Instruct: 30.4 (#210)

| Benchmark | DeepSeek-V2.5 (Sep 2024) | Granite 3.1 8b Instruct |
|---|---|---|
| LMArena Math | 1289 | 1152 |

## Knowledge

- DeepSeek-V2.5 (Sep 2024): 34.2 (#190)
- Granite 3.1 8b Instruct: 30.6 (#218)

| Benchmark | DeepSeek-V2.5 (Sep 2024) | Granite 3.1 8b Instruct |
|---|---|---|
| LMArena Expert | 1266 | 1142 |

## Multilingual

- DeepSeek-V2.5 (Sep 2024): 42.7 (#188)
- Granite 3.1 8b Instruct: 31.1 (#255)

| Benchmark | DeepSeek-V2.5 (Sep 2024) | Granite 3.1 8b Instruct |
|---|---|---|
| LMArena Non-English | 1273 | 1099 |
| LMArena Chinese | 1317 | 1145 |
| LMArena Russian | 1289 | 1092 |
| LMArena French | 1288 | — |
| LMArena German | 1258 | — |
| LMArena Japanese | 1228 | — |
| LMArena Korean | 1209 | — |
| LMArena Spanish | 1248 | — |

## Instruction Following

- DeepSeek-V2.5 (Sep 2024): 67.7 (#189)
- Granite 3.1 8b Instruct: 58.9 (#254)

| Benchmark | DeepSeek-V2.5 (Sep 2024) | Granite 3.1 8b Instruct |
|---|---|---|
| LMArena Instruction Following | 1280 | 1131 |

## Long Context

- DeepSeek-V2.5 (Sep 2024): 39.5 (#169)
- Granite 3.1 8b Instruct: 35.2 (#236)

| Benchmark | DeepSeek-V2.5 (Sep 2024) | Granite 3.1 8b Instruct |
|---|---|---|
| LMArena Longer Query | 1301 | 1162 |

## Writing & Preference

- DeepSeek-V2.5 (Sep 2024): 50.0 (#182)
- Granite 3.1 8b Instruct: 35.7 (#261)

| Benchmark | DeepSeek-V2.5 (Sep 2024) | Granite 3.1 8b Instruct |
|---|---|---|
| LMArena Text | 1294 | 1150 |
| LMArena Creative Writing | 1285 | 1129 |
| LMArena Multi-Turn | 1297 | 1108 |

## FAQ

### Is DeepSeek-V2.5 (Sep 2024) better than Granite 3.1 8b Instruct?

DeepSeek-V2.5 (Sep 2024) is the stronger model overall, scoring 37.0 to 31.8 on the Noometry Index.

### Is DeepSeek-V2.5 (Sep 2024) or Granite 3.1 8b Instruct better for coding?

Granite 3.1 8b Instruct scores higher on coding benchmarks: 33.9 versus 31.3 in the Noometry coding category.

### How many benchmarks do DeepSeek-V2.5 (Sep 2024) and Granite 3.1 8b Instruct share?

12 benchmarks have published results for both models. DeepSeek-V2.5 (Sep 2024) has 22 scored results on Noometry and Granite 3.1 8b Instruct has 13.