Model comparison
Granite 4.0 H Small vs Llama 3-70B
Granite 4.0 H Small is the stronger model overall, scoring 35.9 to 28.5 on the Noometry Index.
Last verified . 13 shared benchmarks.
MetaSummary
- They share 13 benchmarks with published results for both. Granite 4.0 H Small scores higher in 8 categories and Llama 3-70B in 0 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in math, where Granite 4.0 H Small leads 29.3 to 12.0.
Side by side
| Granite 4.0 H Small | Llama 3-70B | |
|---|---|---|
| Provider | IBM | Meta |
| Noometry Index | 35.9 | 28.5 |
| Released | — | 2024-04-18 |
| Weights | Open | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 19 | 31 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
Granite 4.0 H Small: 35.8 (#207), Llama 3-70B: 35.3 (#215)
| Benchmark | Granite 4.0 H Small | Llama 3-70B |
|---|---|---|
| LMArena Coding | 1247 | 1206 |
| BigCodeBench Instruct | — | 43.6% |
| BigCodeBench Complete | — | 54.5% |
| HumanEval+ | — | 72% |
| MBPP+ | — | 69% |
Agentic & Tool Use Not comparable
Granite 4.0 H Small: —, Llama 3-70B: 20.9 (#132)
| Benchmark | Granite 4.0 H Small | Llama 3-70B |
|---|---|---|
| Cybench | — | 5% |
Reasoning Granite 4.0 H Small leads
Granite 4.0 H Small: 23.4 (#164), Llama 3-70B: 17.4 (#284)
| Benchmark | Granite 4.0 H Small | Llama 3-70B |
|---|---|---|
| LMArena Hard Prompts | 1240 | 1195 |
| Kagi LLM Benchmark | — | 35.1% |
| DTBench | — | 54.2% |
| Epoch Capabilities Index | — | 122.91 |
| ForecastBench | — | 57.1 |
| WinoGrande | — | 83.5% |
Math Granite 4.0 H Small leads
Granite 4.0 H Small: 29.3 (#219), Llama 3-70B: 12.0 (#299)
| Benchmark | Granite 4.0 H Small | Llama 3-70B |
|---|---|---|
| LMArena Math | 1251 | 1218 |
| OTIS Mock AIME 2024-2025 | — | 4.3% |
| Omni-MATH | 29.6% | — |
| MATH Level 5 | — | 22.6% |
Knowledge Granite 4.0 H Small leads
Granite 4.0 H Small: 31.5 (#210), Llama 3-70B: 20.8 (#272)
| Benchmark | Granite 4.0 H Small | Llama 3-70B |
|---|---|---|
| LMArena Expert | 1244 | 1149 |
| GPQA Diamond | — | 40.6% |
| MMLU-Pro | 56.9% | — |
| Vectara Hallucination Rate | 5.2% | — |
| GPQA (HELM) | 38.3% | — |
| MMLU | — | 79.3% |
Multilingual Granite 4.0 H Small leads
Granite 4.0 H Small: 38.7 (#223), Llama 3-70B: 33.8 (#246)
| Benchmark | Granite 4.0 H Small | Llama 3-70B |
|---|---|---|
| LMArena Non-English | 1216 | 1142 |
| LMArena Chinese | 1250 | 1114 |
| LMArena Russian | 1198 | 1159 |
| LMArena Spanish | 1258 | 1241 |
| LMArena French | — | 1231 |
| LMArena German | — | 1169 |
| LMArena Japanese | — | 1017 |
| LMArena Korean | — | 1017 |
Instruction Following Granite 4.0 H Small leads
Granite 4.0 H Small: 68.9 (#178), Llama 3-70B: 62.8 (#233)
| Benchmark | Granite 4.0 H Small | Llama 3-70B |
|---|---|---|
| LMArena Instruction Following | 1221 | 1194 |
| IFEval | 89% | — |
Long Context Granite 4.0 H Small leads
Granite 4.0 H Small: 37.6 (#209), Llama 3-70B: 35.6 (#235)
| Benchmark | Granite 4.0 H Small | Llama 3-70B |
|---|---|---|
| LMArena Longer Query | 1240 | 1174 |
Writing & Preference Too close to call
Granite 4.0 H Small: 44.0 (#222), Llama 3-70B: 43.0 (#225)
| Benchmark | Granite 4.0 H Small | Llama 3-70B |
|---|---|---|
| LMArena Text | 1241 | 1221 |
| LMArena Creative Writing | 1210 | 1210 |
| LMArena Multi-Turn | 1243 | 1223 |
| WildBench | 73.9% | — |
Frequently asked questions
Is Granite 4.0 H Small better than Llama 3-70B?
Granite 4.0 H Small is the stronger model overall, scoring 35.9 to 28.5 on the Noometry Index.
Is Granite 4.0 H Small or Llama 3-70B better for coding?
They score almost the same on coding (35.8 vs 35.3); test both on your own repository before choosing.
How many benchmarks do Granite 4.0 H Small and Llama 3-70B share?
13 benchmarks have published results for both models. Granite 4.0 H Small has 19 scored results on Noometry and Llama 3-70B has 31.