Model comparison
Llama 2-7B vs Sonar
Sonar is the stronger model overall, scoring 37.8 to 28.6 on the Noometry Index.
Last verified . 0 shared benchmarks.
Meta
PerplexitySummary
- The widest gap is in writing & preference, where Sonar leads 52.6 to 28.1.
- Llama 2-7B has downloadable open weights; the other is API-only.
Side by side
| Llama 2-7B | Sonar | |
|---|---|---|
| Provider | Meta | Perplexity |
| Noometry Index | 28.6 | 37.8 |
| Released | 2023-07-18 | 2024-01-01 |
| Weights | Open | Proprietary |
| Context window | — | 128K |
| Max output | — | 4K |
| Input $ / M tokens | — | $1 |
| Output $ / M tokens | — | $1 |
| Results tracked | 29 | 7 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Sonar leads
Llama 2-7B: 28.7 (#302), Sonar: 35.1 (#217)
| Benchmark | Llama 2-7B | Sonar |
|---|---|---|
| LiveBench Coding | — | 35.1% |
| LMArena Coding | 1002 | — |
Reasoning Sonar leads
Llama 2-7B: 15.1 (#309), Sonar: 20.3 (#221)
| Benchmark | Llama 2-7B | Sonar |
|---|---|---|
| Chess Puzzles | 0% | — |
| LiveBench Reasoning | — | 46.3% |
| LMArena Hard Prompts | 1009 | — |
| LiveBench Data Analysis | — | 37.9% |
| BIG-Bench Hard | 39.2% | — |
| Epoch Capabilities Index | 98.54 | — |
| HellaSwag | 77.2% | — |
| LAMBADA | 73.3% | — |
| LiveBench | — | 46.9% |
| PIQA | 78.8% | — |
| WinoGrande | 69.2% | — |
Math Sonar leads
Llama 2-7B: 28.2 (#231), Sonar: 31.3 (#197)
| Benchmark | Llama 2-7B | Sonar |
|---|---|---|
| LiveBench Math | — | 41.6% |
| LMArena Math | 1042 | — |
| GSM8K | 16.7% | — |
Knowledge Not comparable
Llama 2-7B: 27.7 (#243), Sonar: —
| Benchmark | Llama 2-7B | Sonar |
|---|---|---|
| LMArena Expert | 1036 | — |
| ARC (AI2) Challenge | 45.9% | — |
| BoolQ | 77.9% | — |
| MMLU | 45.8% | — |
| OpenBookQA | 58.6% | — |
| TriviaQA | 73.7% | — |
Multimodal Not comparable
Llama 2-7B: —, Sonar: —
| Benchmark | Llama 2-7B | Sonar |
|---|---|---|
| ScienceQA | 43.1% | — |
Multilingual Not comparable
Llama 2-7B: 23.9 (#288), Sonar: —
| Benchmark | Llama 2-7B | Sonar |
|---|---|---|
| LMArena Non-English | 973 | — |
| LMArena Chinese | 973 | — |
| LMArena French | 969 | — |
| LMArena German | 978 | — |
| LMArena Russian | 995 | — |
| LMArena Spanish | 1007 | — |
Instruction Following Sonar leads
Llama 2-7B: 51.0 (#293), Sonar: 71.5 (#146)
| Benchmark | Llama 2-7B | Sonar |
|---|---|---|
| LiveBench Instruction Following | — | 76.2% |
| LMArena Instruction Following | 1006 | — |
Long Context Not comparable
Llama 2-7B: 30.4 (#282), Sonar: —
| Benchmark | Llama 2-7B | Sonar |
|---|---|---|
| LMArena Longer Query | 999 | — |
Writing & Preference Sonar leads
Llama 2-7B: 28.1 (#293), Sonar: 52.6 (#166)
| Benchmark | Llama 2-7B | Sonar |
|---|---|---|
| LMArena Text | 1053 | — |
| LMArena Creative Writing | 1033 | — |
| LMArena Multi-Turn | 1029 | — |
| LiveBench Language | — | 44.1% |
Frequently asked questions
Is Llama 2-7B better than Sonar?
Sonar is the stronger model overall, scoring 37.8 to 28.6 on the Noometry Index.
Is Llama 2-7B or Sonar better for coding?
Sonar scores higher on coding benchmarks: 35.1 versus 28.7 in the Noometry coding category.
How many benchmarks do Llama 2-7B and Sonar share?
0 benchmarks have published results for both models. Llama 2-7B has 29 scored results on Noometry and Sonar has 7.