# GPT-5.4 mini vs Granite 3.1 8b Instruct

> GPT-5.4 mini is the stronger model overall, scoring 44.8 to 31.8 on the Noometry Index.

- Canonical page: https://noometry.com/compare/gpt-5-4-mini-vs-granite-3-1-8b-instruct
- Last updated: 2026-10-10
- Shared benchmarks: 12

## Summary

- They share 12 benchmarks with published results for both. GPT-5.4 mini scores higher in 9 categories and Granite 3.1 8b Instruct in 0 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where GPT-5.4 mini leads 64.2 to 35.7.
- Granite 3.1 8b Instruct has downloadable open weights; the other is API-only.

## Snapshot

| | GPT-5.4 mini | Granite 3.1 8b Instruct |
|---|---|---|
| Provider | OpenAI | IBM |
| Noometry Index | 44.8 | 31.8 |
| Rank | 69 | 256 |
| Context | 400K | — |
| Input $/M | $0.75 | — |
| Output $/M | $4.50 | — |
| Weights | Proprietary | Open |

## Coding

- GPT-5.4 mini: 45.0 (#67)
- Granite 3.1 8b Instruct: 33.9 (#232)

| Benchmark | GPT-5.4 mini | Granite 3.1 8b Instruct |
|---|---|---|
| LMArena Coding | 1438 | 1186 |
| FrontierCode | 27% | — |
| LMArena WebDev | 1397 | — |
| SciCode | 49.9% | — |
| WeirdML | 60.3% | — |
| ALE-Bench | 1,189 | — |

## Agentic & Tool Use

- GPT-5.4 mini: 29.6 (#74)
- Granite 3.1 8b Instruct: 23.9 (#113)

| Benchmark | GPT-5.4 mini | Granite 3.1 8b Instruct |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 27.1% |
| DeepResearch Bench | 36.3% | — |

## Reasoning

- GPT-5.4 mini: 30.3 (#77)
- Granite 3.1 8b Instruct: 21.2 (#204)

| Benchmark | GPT-5.4 mini | Granite 3.1 8b Instruct |
|---|---|---|
| LMArena Hard Prompts | 1424 | 1146 |
| ARC-AGI-2 | 18.9% | — |
| Kagi LLM Benchmark | 37.9% | — |
| NYT Connections (extended) | 61.8% | — |
| ARC-AGI-1 | 63.7% | — |
| CritPt | 10% | — |
| Chess Puzzles | 24% | — |
| Thematic Generalization | 61.7% | — |
| Mystery Game Puzzles | 11% | — |
| DTBench | 80% | — |
| LMCA | 40.8% | — |
| Epoch Capabilities Index | 148.83 | — |
| ForecastBench | 57 | — |

## Math

- GPT-5.4 mini: 45.4 (#63)
- Granite 3.1 8b Instruct: 30.4 (#210)

| Benchmark | GPT-5.4 mini | Granite 3.1 8b Instruct |
|---|---|---|
| LMArena Math | 1418 | 1152 |
| FrontierMath (Tiers 1-3) | 51.2% | — |
| FrontierMath Tier 4 | 9.8% | — |
| OTIS Mock AIME 2024-2025 | 88.9% | — |
| ProofBench | 21% | — |
| FrontierMath (Feb 2025 set) | 28.3% | — |
| FrontierMath Tier 4 (v1) | 2.1% | — |

## Knowledge

- GPT-5.4 mini: 51.4 (#62)
- Granite 3.1 8b Instruct: 30.6 (#218)

| Benchmark | GPT-5.4 mini | Granite 3.1 8b Instruct |
|---|---|---|
| LMArena Expert | 1434 | 1142 |
| GPQA Diamond | 86.9% | — |
| SimpleQA Verified | 29.4% | — |
| Vectara Hallucination Rate | 5.5% | — |

## Multimodal

- GPT-5.4 mini: 39.0 (#51)
- Granite 3.1 8b Instruct: —

| Benchmark | GPT-5.4 mini | Granite 3.1 8b Instruct |
|---|---|---|
| LMArena Vision | 1245 | — |

## Multilingual

- GPT-5.4 mini: 52.1 (#94)
- Granite 3.1 8b Instruct: 31.1 (#255)

| Benchmark | GPT-5.4 mini | Granite 3.1 8b Instruct |
|---|---|---|
| LMArena Non-English | 1405 | 1099 |
| LMArena Chinese | 1446 | 1145 |
| LMArena Russian | 1417 | 1092 |
| LMArena French | 1440 | — |
| LMArena German | 1407 | — |
| LMArena Japanese | 1375 | — |
| LMArena Korean | 1369 | — |
| LMArena Spanish | 1403 | — |

## Instruction Following

- GPT-5.4 mini: 74.3 (#99)
- Granite 3.1 8b Instruct: 58.9 (#254)

| Benchmark | GPT-5.4 mini | Granite 3.1 8b Instruct |
|---|---|---|
| LMArena Instruction Following | 1405 | 1131 |

## Long Context

- GPT-5.4 mini: 42.9 (#107)
- Granite 3.1 8b Instruct: 35.2 (#236)

| Benchmark | GPT-5.4 mini | Granite 3.1 8b Instruct |
|---|---|---|
| LMArena Longer Query | 1407 | 1162 |

## Writing & Preference

- GPT-5.4 mini: 64.2 (#56)
- Granite 3.1 8b Instruct: 35.7 (#261)

| Benchmark | GPT-5.4 mini | Granite 3.1 8b Instruct |
|---|---|---|
| LMArena Text | 1411 | 1150 |
| LMArena Creative Writing | 1370 | 1129 |
| LMArena Multi-Turn | 1429 | 1108 |
| EQ-Bench Creative Writing | 1665 | — |

## FAQ

### Is GPT-5.4 mini better than Granite 3.1 8b Instruct?

GPT-5.4 mini is the stronger model overall, scoring 44.8 to 31.8 on the Noometry Index.

### Is GPT-5.4 mini or Granite 3.1 8b Instruct better for coding?

GPT-5.4 mini scores higher on coding benchmarks: 45.0 versus 33.9 in the Noometry coding category.

### How many benchmarks do GPT-5.4 mini and Granite 3.1 8b Instruct share?

12 benchmarks have published results for both models. GPT-5.4 mini has 46 scored results on Noometry and Granite 3.1 8b Instruct has 13.