Model comparison
Llama 3.1-8B vs Llama 3-8B
Llama 3-8B is the stronger model overall, scoring 25.5 to 23.0 on the Noometry Index.
Last verified . 28 shared benchmarks.
MetaSummary
- They share 28 benchmarks with published results for both. Llama 3.1-8B scores higher in 6 categories and Llama 3-8B in 2 categories; 5 gaps are clear of the uncertainty.
- The widest gap is in coding, where Llama 3-8B leads 31.0 to 20.2.
- The biggest single-benchmark swing is MATH Level 5: 22.9% for Llama 3.1-8B and 6.1% for Llama 3-8B.
Side by side
| Llama 3.1-8B | Llama 3-8B | |
|---|---|---|
| Provider | Meta | Meta |
| Noometry Index | 23.0 | 25.5 |
| Released | 2024-07-23 | 2024-04-18 |
| Weights | Open | Open |
| Context window | 128K | — |
| Max output | 4K | — |
| Input $ / M tokens | $0.05 | — |
| Output $ / M tokens | $0.08 | — |
| Results tracked | 43 | 34 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Llama 3-8B leads
Llama 3.1-8B: 20.2 (#340), Llama 3-8B: 31.0 (#289)
| Benchmark | Llama 3.1-8B | Llama 3-8B |
|---|---|---|
| BigCodeBench Instruct | 32.8% | 31.9% |
| LMArena Coding | 1195 | 1152 |
| BigCodeBench Complete | 40.5% | 36.9% |
| HumanEval+ | 62.8% | 56.7% |
| MBPP+ | 55.6% | 54.8% |
| SciCode | 13.2% | — |
| WeirdML | 1.7% | — |
Agentic & Tool Use Not comparable
Llama 3.1-8B: 22.5 (#131), Llama 3-8B: —
| Benchmark | Llama 3.1-8B | Llama 3-8B |
|---|---|---|
| Berkeley Function Calling Leaderboard | 25.8% | — |
| BALROG | 15.1% | — |
Reasoning Too close to call
Llama 3.1-8B: 14.9 (#321), Llama 3-8B: 14.3 (#326)
| Benchmark | Llama 3.1-8B | Llama 3-8B |
|---|---|---|
| Chess Puzzles | 0% | 0% |
| LMArena Hard Prompts | 1175 | 1133 |
| DTBench | 50.9% | 43.9% |
| Epoch Capabilities Index | 116.57 | 116.45 |
| CritPt | 0% | — |
| LMCA | 5.4% | — |
| Adversarial NLI | — | 57.3% |
| ForecastBench | — | 58.6 |
| PIQA | 81.2% | — |
| WinoGrande | — | 75.7% |
Math Llama 3.1-8B leads
Llama 3.1-8B: 10.2 (#317), Llama 3-8B: 8.8 (#323)
| Benchmark | Llama 3.1-8B | Llama 3-8B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 1.7% | 1.9% |
| LMArena Math | 1179 | 1151 |
| MATH Level 5 | 22.9% | 6.1% |
| Omni-MATH | 13.7% | — |
| GSM8K | 82.4% | — |
Knowledge Too close to call
Llama 3.1-8B: 8.0 (#307), Llama 3-8B: 7.8 (#308)
| Benchmark | Llama 3.1-8B | Llama 3-8B |
|---|---|---|
| GPQA Diamond | 27% | 26.1% |
| LMArena Expert | 1144 | 1113 |
| MMLU | 56.1% | 68.8% |
| MMLU-Pro | 40.6% | — |
| GPQA (HELM) | 24.7% | — |
| ARC (AI2) Challenge | — | 82.8% |
| BoolQ | 82.8% | — |
| OpenBookQA | — | 82.6% |
| TriviaQA | — | 67.7% |
Multilingual Llama 3.1-8B leads
Llama 3.1-8B: 34.0 (#249), Llama 3-8B: 30.8 (#261)
| Benchmark | Llama 3.1-8B | Llama 3-8B |
|---|---|---|
| LMArena Non-English | 1148 | 1098 |
| LMArena Chinese | 1151 | 1076 |
| LMArena French | 1177 | 1159 |
| LMArena German | 1144 | 1104 |
| LMArena Japanese | 1061 | 967 |
| LMArena Korean | 1053 | 1004 |
| LMArena Russian | 1158 | 1109 |
| LMArena Spanish | 1169 | 1173 |
Instruction Following Too close to call
Llama 3.1-8B: 58.9 (#258), Llama 3-8B: 58.4 (#260)
| Benchmark | Llama 3.1-8B | Llama 3-8B |
|---|---|---|
| LMArena Instruction Following | 1159 | 1127 |
| IFEval | 74.3% | — |
Long Context Llama 3.1-8B leads
Llama 3.1-8B: 35.8 (#238), Llama 3-8B: 34.2 (#251)
| Benchmark | Llama 3.1-8B | Llama 3-8B |
|---|---|---|
| LMArena Longer Query | 1182 | 1128 |
Writing & Preference Llama 3-8B leads
Llama 3.1-8B: 29.7 (#290), Llama 3-8B: 37.5 (#256)
| Benchmark | Llama 3.1-8B | Llama 3-8B |
|---|---|---|
| LMArena Text | 1187 | 1166 |
| LMArena Creative Writing | 1154 | 1150 |
| LMArena Multi-Turn | 1172 | 1152 |
| EQ-Bench Creative Writing | 713 | — |
| WildBench | 68.7% | — |
Frequently asked questions
Is Llama 3.1-8B better than Llama 3-8B?
Llama 3-8B is the stronger model overall, scoring 25.5 to 23.0 on the Noometry Index.
Is Llama 3.1-8B or Llama 3-8B better for coding?
Llama 3-8B scores higher on coding benchmarks: 31.0 versus 20.2 in the Noometry coding category.
How many benchmarks do Llama 3.1-8B and Llama 3-8B share?
28 benchmarks have published results for both models. Llama 3.1-8B has 43 scored results on Noometry and Llama 3-8B has 34.