Model comparison
Granite 4.0 Micro vs Llama 2-70B
Granite 4.0 Micro is the stronger model overall, scoring 29.0 to 24.4 on the Noometry Index.
Last verified . 2 shared benchmarks.
MetaSummary
- They share 2 benchmarks with published results for both. Granite 4.0 Micro scores higher in 5 categories and Llama 2-70B in 0 categories; 5 gaps are clear of the uncertainty.
- The widest gap is in instruction following, where Granite 4.0 Micro leads 69.9 to 54.9.
Side by side
| Granite 4.0 Micro | Llama 2-70B | |
|---|---|---|
| Provider | IBM | Meta |
| Noometry Index | 29.0 | 24.4 |
| Released | 2025-10-02 | 2023-07-18 |
| Weights | Open | Open |
| Context window | 131K | — |
| Max output | 118K | — |
| Input $ / M tokens | $0.017 | — |
| Output $ / M tokens | $0.11 | — |
| Results tracked | 8 | 35 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Not comparable
Granite 4.0 Micro: —, Llama 2-70B: 31.4 (#286)
| Benchmark | Granite 4.0 Micro | Llama 2-70B |
|---|---|---|
| LMArena Coding | — | 1079 |
Reasoning Granite 4.0 Micro leads
Granite 4.0 Micro: 19.2 (#265), Llama 2-70B: 14.4 (#325)
| Benchmark | Granite 4.0 Micro | Llama 2-70B |
|---|---|---|
| Chess Puzzles | 0% | — |
| LMArena Hard Prompts | — | 1073 |
| DTBench | — | 41.6% |
| BIG-Bench Hard | — | 64.9% |
| CommonsenseQA 2.0 | — | 50% |
| Epoch Capabilities Index | — | 113.79 |
| ForecastBench | — | 51.4 |
| HellaSwag | — | 85.3% |
| LAMBADA | — | 78.9% |
| PIQA | — | 82.8% |
| WinoGrande | — | 80.2% |
Math Granite 4.0 Micro leads
Granite 4.0 Micro: 12.0 (#307), Llama 2-70B: 8.1 (#326)
| Benchmark | Granite 4.0 Micro | Llama 2-70B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 2.8% | 0% |
| Omni-MATH | 20.9% | — |
| LMArena Math | — | 1091 |
| MATH Level 5 | — | 3.3% |
| GSM8K | — | 69.6% |
Knowledge Granite 4.0 Micro leads
Granite 4.0 Micro: 9.9 (#304), Llama 2-70B: 7.4 (#310)
| Benchmark | Granite 4.0 Micro | Llama 2-70B |
|---|---|---|
| GPQA Diamond | 28.3% | 26.3% |
| MMLU-Pro | 39.5% | — |
| GPQA (HELM) | 30.7% | — |
| LMArena Expert | — | 1039 |
| ARC (AI2) Challenge | — | 78.3% |
| BoolQ | — | 88.6% |
| MMLU | — | 69.9% |
| OpenBookQA | — | 60.2% |
| TriviaQA | — | 87.6% |
Multilingual Not comparable
Granite 4.0 Micro: —, Llama 2-70B: 27.7 (#274)
| Benchmark | Granite 4.0 Micro | Llama 2-70B |
|---|---|---|
| LMArena Non-English | — | 1045 |
| LMArena Chinese | — | 995 |
| LMArena French | — | 1090 |
| LMArena German | — | 1041 |
| LMArena Japanese | — | 927 |
| LMArena Korean | — | 964 |
| LMArena Russian | — | 1083 |
| LMArena Spanish | — | 1143 |
Instruction Following Granite 4.0 Micro leads
Granite 4.0 Micro: 69.9 (#169), Llama 2-70B: 54.9 (#278)
| Benchmark | Granite 4.0 Micro | Llama 2-70B |
|---|---|---|
| IFEval | 84.9% | — |
| LMArena Instruction Following | — | 1071 |
Long Context Not comparable
Granite 4.0 Micro: —, Llama 2-70B: 32.3 (#270)
| Benchmark | Granite 4.0 Micro | Llama 2-70B |
|---|---|---|
| LMArena Longer Query | — | 1062 |
Writing & Preference Granite 4.0 Micro leads
Granite 4.0 Micro: 46.7 (#216), Llama 2-70B: 32.3 (#279)
| Benchmark | Granite 4.0 Micro | Llama 2-70B |
|---|---|---|
| LMArena Text | — | 1115 |
| LMArena Creative Writing | — | 1075 |
| WildBench | 67% | — |
| LMArena Multi-Turn | — | 1088 |
Frequently asked questions
Is Granite 4.0 Micro better than Llama 2-70B?
Granite 4.0 Micro is the stronger model overall, scoring 29.0 to 24.4 on the Noometry Index.
How many benchmarks do Granite 4.0 Micro and Llama 2-70B share?
2 benchmarks have published results for both models. Granite 4.0 Micro has 8 scored results on Noometry and Llama 2-70B has 35.