# Granite 4.0 Micro vs Trinity Large Thinking

> Trinity Large Thinking is the stronger model overall, scoring 38.6 to 29.0 on the Noometry Index. Granite 4.0 Micro costs 9.5× less per token, which makes it the better buy when Trinity Large Thinking's lead doesn't matter for your workload.

- Canonical page: https://noometry.com/compare/granite-4-0-micro-vs-trinity-large-thinking
- Last updated: 2026-10-11
- Shared benchmarks: 0

## Summary

- The widest gap is in knowledge, where Trinity Large Thinking leads 40.9 to 9.9.
- Granite 4.0 Micro is cheaper at $0.017 / $0.11 per million input/output tokens, against $0.25 / $0.80 for Trinity Large Thinking.
- Trinity Large Thinking accepts more context: 262K tokens versus 131K.

## Snapshot

| | Granite 4.0 Micro | Trinity Large Thinking |
|---|---|---|
| Provider | IBM | Arcee AI |
| Noometry Index | 29.0 | 38.6 |
| Rank | 318 | 185 |
| Context | 131K | 262K |
| Input $/M | $0.017 | $0.25 |
| Output $/M | $0.11 | $0.80 |
| Weights | Open | Open |

## Coding

- Granite 4.0 Micro: —
- Trinity Large Thinking: 34.1 (#244)

| Benchmark | Granite 4.0 Micro | Trinity Large Thinking |
|---|---|---|
| LMArena WebDev | — | 1238 |
| SciCode | — | 36.1% |
| LMArena Coding | — | 1381 |

## Reasoning

- Granite 4.0 Micro: 19.2 (#265)
- Trinity Large Thinking: 16.9 (#298)

| Benchmark | Granite 4.0 Micro | Trinity Large Thinking |
|---|---|---|
| NYT Connections (extended) | — | 16.5% |
| CritPt | — | 0.9% |
| Chess Puzzles | 0% | — |
| Thematic Generalization | — | 41.6% |
| LMArena Hard Prompts | — | 1350 |
| Surface Evolver Bench | — | 15.6% |

## Math

- Granite 4.0 Micro: 12.0 (#307)
- Trinity Large Thinking: 37.6 (#149)

| Benchmark | Granite 4.0 Micro | Trinity Large Thinking |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 2.8% | — |
| Omni-MATH | 20.9% | — |
| LMArena Math | — | 1366 |

## Knowledge

- Granite 4.0 Micro: 9.9 (#304)
- Trinity Large Thinking: 40.9 (#113)

| Benchmark | Granite 4.0 Micro | Trinity Large Thinking |
|---|---|---|
| GPQA Diamond | 28.3% | — |
| MMLU-Pro | 39.5% | — |
| Vectara Hallucination Rate | — | 6.9% |
| GPQA (HELM) | 30.7% | — |
| LMArena Expert | — | 1360 |

## Multilingual

- Granite 4.0 Micro: —
- Trinity Large Thinking: 46.2 (#160)

| Benchmark | Granite 4.0 Micro | Trinity Large Thinking |
|---|---|---|
| LMArena Non-English | — | 1325 |
| LMArena Chinese | — | 1373 |
| LMArena French | — | 1374 |
| LMArena German | — | 1356 |
| LMArena Japanese | — | 1311 |
| LMArena Korean | — | 1306 |
| LMArena Russian | — | 1337 |
| LMArena Spanish | — | 1357 |

## Instruction Following

- Granite 4.0 Micro: 69.9 (#169)
- Trinity Large Thinking: 70.5 (#162)

| Benchmark | Granite 4.0 Micro | Trinity Large Thinking |
|---|---|---|
| IFEval | 84.9% | — |
| LMArena Instruction Following | — | 1334 |

## Long Context

- Granite 4.0 Micro: —
- Trinity Large Thinking: 41.3 (#144)

| Benchmark | Granite 4.0 Micro | Trinity Large Thinking |
|---|---|---|
| LMArena Longer Query | — | 1355 |

## Writing & Preference

- Granite 4.0 Micro: 46.7 (#216)
- Trinity Large Thinking: 53.8 (#158)

| Benchmark | Granite 4.0 Micro | Trinity Large Thinking |
|---|---|---|
| LMArena Text | — | 1340 |
| LMArena Creative Writing | — | 1320 |
| WildBench | 67% | — |
| LMArena Multi-Turn | — | 1342 |

## FAQ

### Is Granite 4.0 Micro better than Trinity Large Thinking?

Trinity Large Thinking is the stronger model overall, scoring 38.6 to 29.0 on the Noometry Index. Granite 4.0 Micro costs 9.5× less per token, which makes it the better buy when Trinity Large Thinking's lead doesn't matter for your workload.

### Which is cheaper, Granite 4.0 Micro or Trinity Large Thinking?

Granite 4.0 Micro is cheaper. It lists at $0.017 per million input tokens and $0.11 per million output tokens; Trinity Large Thinking lists at $0.25 and $0.80.

### Which has the bigger context window?

Trinity Large Thinking does, with 262K tokens against 131K.

### How many benchmarks do Granite 4.0 Micro and Trinity Large Thinking share?

0 benchmarks have published results for both models. Granite 4.0 Micro has 8 scored results on Noometry and Trinity Large Thinking has 24.