Model comparison
Granite 3.0 2b Instruct vs Llama 3-8B
Granite 3.0 2b Instruct is the stronger model overall, scoring 30.8 to 25.5 on the Noometry Index.
Last verified . 13 shared benchmarks.
MetaSummary
- They share 13 benchmarks with published results for both. Granite 3.0 2b Instruct scores higher in 3 categories and Llama 3-8B in 5 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Granite 3.0 2b Instruct leads 32.2 to 8.8.
- The biggest single-benchmark swing is BigCodeBench Instruct: 20.5% for Granite 3.0 2b Instruct and 31.9% for Llama 3-8B.
Side by side
| Granite 3.0 2b Instruct | Llama 3-8B | |
|---|---|---|
| Provider | IBM | Meta |
| Noometry Index | 30.8 | 25.5 |
| Released | — | 2024-04-18 |
| Weights | Open | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 13 | 34 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Llama 3-8B leads
Granite 3.0 2b Instruct: 28.3 (#316), Llama 3-8B: 31.0 (#289)
| Benchmark | Granite 3.0 2b Instruct | Llama 3-8B |
|---|---|---|
| BigCodeBench Instruct | 20.5% | 31.9% |
| LMArena Coding | 1090 | 1152 |
| BigCodeBench Complete | — | 36.9% |
| HumanEval+ | — | 56.7% |
| MBPP+ | — | 54.8% |
Reasoning Granite 3.0 2b Instruct leads
Granite 3.0 2b Instruct: 20.5 (#235), Llama 3-8B: 14.3 (#326)
| Benchmark | Granite 3.0 2b Instruct | Llama 3-8B |
|---|---|---|
| LMArena Hard Prompts | 1073 | 1133 |
| Chess Puzzles | — | 0% |
| DTBench | — | 43.9% |
| Adversarial NLI | — | 57.3% |
| Epoch Capabilities Index | — | 116.45 |
| ForecastBench | — | 58.6 |
| WinoGrande | — | 75.7% |
Math Granite 3.0 2b Instruct leads
Granite 3.0 2b Instruct: 32.2 (#217), Llama 3-8B: 8.8 (#323)
| Benchmark | Granite 3.0 2b Instruct | Llama 3-8B |
|---|---|---|
| LMArena Math | 1117 | 1151 |
| OTIS Mock AIME 2024-2025 | — | 1.9% |
| MATH Level 5 | — | 6.1% |
Knowledge Granite 3.0 2b Instruct leads
Granite 3.0 2b Instruct: 29.0 (#241), Llama 3-8B: 7.8 (#308)
| Benchmark | Granite 3.0 2b Instruct | Llama 3-8B |
|---|---|---|
| LMArena Expert | 1064 | 1113 |
| GPQA Diamond | — | 26.1% |
| ARC (AI2) Challenge | — | 82.8% |
| MMLU | — | 68.8% |
| OpenBookQA | — | 82.6% |
| TriviaQA | — | 67.7% |
Multilingual Llama 3-8B leads
Granite 3.0 2b Instruct: 27.0 (#278), Llama 3-8B: 30.8 (#261)
| Benchmark | Granite 3.0 2b Instruct | Llama 3-8B |
|---|---|---|
| LMArena Non-English | 1033 | 1098 |
| LMArena Chinese | 1070 | 1076 |
| LMArena Russian | 1045 | 1109 |
| LMArena French | — | 1159 |
| LMArena German | — | 1104 |
| LMArena Japanese | — | 967 |
| LMArena Korean | — | 1004 |
| LMArena Spanish | — | 1173 |
Instruction Following Llama 3-8B leads
Granite 3.0 2b Instruct: 53.9 (#284), Llama 3-8B: 58.4 (#260)
| Benchmark | Granite 3.0 2b Instruct | Llama 3-8B |
|---|---|---|
| LMArena Instruction Following | 1056 | 1127 |
Long Context Llama 3-8B leads
Granite 3.0 2b Instruct: 32.5 (#268), Llama 3-8B: 34.2 (#251)
| Benchmark | Granite 3.0 2b Instruct | Llama 3-8B |
|---|---|---|
| LMArena Longer Query | 1070 | 1128 |
Writing & Preference Llama 3-8B leads
Granite 3.0 2b Instruct: 29.6 (#292), Llama 3-8B: 37.5 (#256)
| Benchmark | Granite 3.0 2b Instruct | Llama 3-8B |
|---|---|---|
| LMArena Text | 1080 | 1166 |
| LMArena Creative Writing | 1046 | 1150 |
| LMArena Multi-Turn | 1053 | 1152 |
Frequently asked questions
Is Granite 3.0 2b Instruct better than Llama 3-8B?
Granite 3.0 2b Instruct is the stronger model overall, scoring 30.8 to 25.5 on the Noometry Index.
Is Granite 3.0 2b Instruct or Llama 3-8B better for coding?
Llama 3-8B scores higher on coding benchmarks: 31.0 versus 28.3 in the Noometry coding category.
How many benchmarks do Granite 3.0 2b Instruct and Llama 3-8B share?
13 benchmarks have published results for both models. Granite 3.0 2b Instruct has 13 scored results on Noometry and Llama 3-8B has 34.