# Granite 4.0 Micro vs Llama 2-34B

> Granite 4.0 Micro has enough public results to be ranked (#312); Llama 2-34B does not yet, so treat this comparison as directional.

- Canonical page: https://noometry.com/compare/granite-4-0-micro-vs-llama-2-34b
- Last updated: 2026-10-10
- Shared benchmarks: 0

## Summary

- Granite 4.0 Micro has downloadable open weights; the other is API-only.

## Snapshot

| | Granite 4.0 Micro | Llama 2-34B |
|---|---|---|
| Provider | IBM | Meta |
| Noometry Index | 28.8 | — |
| Rank | 312 | — |
| Context | 131K | — |
| Input $/M | $0.017 | — |
| Output $/M | $0.11 | — |
| Weights | Open | Proprietary |

## Reasoning

- Granite 4.0 Micro: 18.5 (#261)
- Llama 2-34B: —

| Benchmark | Granite 4.0 Micro | Llama 2-34B |
|---|---|---|
| Chess Puzzles | 0% | — |
| BIG-Bench Hard | — | 44.1% |
| Epoch Capabilities Index | — | 104.81 |
| PIQA | — | 81.9% |
| WinoGrande | — | 76.7% |

## Math

- Granite 4.0 Micro: 11.2 (#301)
- Llama 2-34B: —

| Benchmark | Granite 4.0 Micro | Llama 2-34B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 2.8% | — |
| Omni-MATH | 20.9% | — |
| GSM8K | — | 42.2% |

## Knowledge

- Granite 4.0 Micro: 9.9 (#299)
- Llama 2-34B: —

| Benchmark | Granite 4.0 Micro | Llama 2-34B |
|---|---|---|
| GPQA Diamond | 28.3% | — |
| MMLU-Pro | 39.5% | — |
| GPQA (HELM) | 30.7% | — |
| ARC (AI2) Challenge | — | 54.5% |
| BoolQ | — | 83.7% |
| MMLU | — | 62.6% |
| OpenBookQA | — | 58.2% |
| TriviaQA | — | 84.6% |

## Instruction Following

- Granite 4.0 Micro: 70.0 (#164)
- Llama 2-34B: —

| Benchmark | Granite 4.0 Micro | Llama 2-34B |
|---|---|---|
| IFEval | 84.9% | — |

## Writing & Preference

- Granite 4.0 Micro: 46.7 (#212)
- Llama 2-34B: —

| Benchmark | Granite 4.0 Micro | Llama 2-34B |
|---|---|---|
| WildBench | 67% | — |

## FAQ

### Is Granite 4.0 Micro better than Llama 2-34B?

Granite 4.0 Micro has enough public results to be ranked (#312); Llama 2-34B does not yet, so treat this comparison as directional.

### How many benchmarks do Granite 4.0 Micro and Llama 2-34B share?

0 benchmarks have published results for both models. Granite 4.0 Micro has 8 scored results on Noometry and Llama 2-34B has 10.