Model comparison
Llama 3-8B vs Sonar
Sonar is the stronger model overall, scoring 38.5 to 25.5 on the Noometry Index.
Last verified . 0 shared benchmarks.
Meta
PerplexitySummary
- The widest gap is in math, where Sonar leads 33.7 to 8.8.
- Llama 3-8B has downloadable open weights; the other is API-only.
Side by side
| Llama 3-8B | Sonar | |
|---|---|---|
| Provider | Meta | Perplexity |
| Noometry Index | 25.5 | 38.5 |
| Released | 2024-04-18 | 2024-01-01 |
| Weights | Open | Proprietary |
| Context window | — | 128K |
| Max output | — | 4K |
| Input $ / M tokens | — | $1 |
| Output $ / M tokens | — | $1 |
| Results tracked | 34 | 7 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Sonar leads
Llama 3-8B: 31.0 (#289), Sonar: 35.7 (#221)
| Benchmark | Llama 3-8B | Sonar |
|---|---|---|
| BigCodeBench Instruct | 31.9% | — |
| LiveBench Coding | — | 35.1% |
| LMArena Coding | 1152 | — |
| BigCodeBench Complete | 36.9% | — |
| HumanEval+ | 56.7% | — |
| MBPP+ | 54.8% | — |
Reasoning Sonar leads
Llama 3-8B: 14.3 (#326), Sonar: 21.1 (#227)
| Benchmark | Llama 3-8B | Sonar |
|---|---|---|
| Chess Puzzles | 0% | — |
| LiveBench Reasoning | — | 46.3% |
| LMArena Hard Prompts | 1133 | — |
| DTBench | 43.9% | — |
| LiveBench Data Analysis | — | 37.9% |
| Adversarial NLI | 57.3% | — |
| Epoch Capabilities Index | 116.45 | — |
| ForecastBench | 58.6 | — |
| LiveBench | — | 46.9% |
| WinoGrande | 75.7% | — |
Math Sonar leads
Llama 3-8B: 8.8 (#323), Sonar: 33.7 (#200)
| Benchmark | Llama 3-8B | Sonar |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 1.9% | — |
| LiveBench Math | — | 41.6% |
| LMArena Math | 1151 | — |
| MATH Level 5 | 6.1% | — |
Knowledge Not comparable
Llama 3-8B: 7.8 (#308), Sonar: —
| Benchmark | Llama 3-8B | Sonar |
|---|---|---|
| GPQA Diamond | 26.1% | — |
| LMArena Expert | 1113 | — |
| ARC (AI2) Challenge | 82.8% | — |
| MMLU | 68.8% | — |
| OpenBookQA | 82.6% | — |
| TriviaQA | 67.7% | — |
Multilingual Not comparable
Llama 3-8B: 30.8 (#261), Sonar: —
| Benchmark | Llama 3-8B | Sonar |
|---|---|---|
| LMArena Non-English | 1098 | — |
| LMArena Chinese | 1076 | — |
| LMArena French | 1159 | — |
| LMArena German | 1104 | — |
| LMArena Japanese | 967 | — |
| LMArena Korean | 1004 | — |
| LMArena Russian | 1109 | — |
| LMArena Spanish | 1173 | — |
Instruction Following Sonar leads
Llama 3-8B: 58.4 (#260), Sonar: 71.4 (#150)
| Benchmark | Llama 3-8B | Sonar |
|---|---|---|
| LiveBench Instruction Following | — | 76.2% |
| LMArena Instruction Following | 1127 | — |
Long Context Not comparable
Llama 3-8B: 34.2 (#251), Sonar: —
| Benchmark | Llama 3-8B | Sonar |
|---|---|---|
| LMArena Longer Query | 1128 | — |
Writing & Preference Sonar leads
Llama 3-8B: 37.5 (#256), Sonar: 52.6 (#167)
| Benchmark | Llama 3-8B | Sonar |
|---|---|---|
| LMArena Text | 1166 | — |
| LMArena Creative Writing | 1150 | — |
| LMArena Multi-Turn | 1152 | — |
| LiveBench Language | — | 44.1% |
Frequently asked questions
Is Llama 3-8B better than Sonar?
Sonar is the stronger model overall, scoring 38.5 to 25.5 on the Noometry Index.
Is Llama 3-8B or Sonar better for coding?
Sonar scores higher on coding benchmarks: 35.7 versus 31.0 in the Noometry coding category.
How many benchmarks do Llama 3-8B and Sonar share?
0 benchmarks have published results for both models. Llama 3-8B has 34 scored results on Noometry and Sonar has 7.