Model comparison
Llama 2-70B vs Solar Pro4
Solar Pro4 is the stronger model overall, scoring 41.5 to 24.1 on the Noometry Index.
Last verified . 17 shared benchmarks.
Meta
UpstageSummary
- They share 17 benchmarks with published results for both. Llama 2-70B scores higher in 0 categories and Solar Pro4 in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Solar Pro4 leads 39.0 to 7.4.
- Llama 2-70B has downloadable open weights; the other is API-only.
Side by side
| Llama 2-70B | Solar Pro4 | |
|---|---|---|
| Provider | Meta | Upstage |
| Noometry Index | 24.1 | 41.5 |
| Released | 2023-07-18 | 2026-08-06 |
| Weights | Open | Proprietary |
| Context window | — | 524K |
| Max output | — | 131K |
| Input $ / M tokens | — | $0.30 |
| Output $ / M tokens | — | $1.20 |
| Results tracked | 35 | 18 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Solar Pro4 leads
Llama 2-70B: 30.8 (#281), Solar Pro4: 39.9 (#135)
| Benchmark | Llama 2-70B | Solar Pro4 |
|---|---|---|
| LMArena Coding | 1079 | 1439 |
| LMArena WebDev | — | 1371 |
Reasoning Solar Pro4 leads
Llama 2-70B: 13.7 (#324), Solar Pro4: 27.3 (#107)
| Benchmark | Llama 2-70B | Solar Pro4 |
|---|---|---|
| LMArena Hard Prompts | 1073 | 1398 |
| DTBench | 41.6% | — |
| BIG-Bench Hard | 64.9% | — |
| CommonsenseQA 2.0 | 50% | — |
| Epoch Capabilities Index | 113.59 | — |
| ForecastBench | 51.4 | — |
| HellaSwag | 85.3% | — |
| LAMBADA | 78.9% | — |
| PIQA | 82.8% | — |
| WinoGrande | 80.2% | — |
Math Solar Pro4 leads
Llama 2-70B: 7.5 (#320), Solar Pro4: 36.1 (#129)
| Benchmark | Llama 2-70B | Solar Pro4 |
|---|---|---|
| LMArena Math | 1091 | 1418 |
| OTIS Mock AIME 2024-2025 | 0% | — |
| MATH Level 5 | 3.3% | — |
| GSM8K | 69.6% | — |
Knowledge Solar Pro4 leads
Llama 2-70B: 7.4 (#305), Solar Pro4: 39.0 (#128)
| Benchmark | Llama 2-70B | Solar Pro4 |
|---|---|---|
| LMArena Expert | 1039 | 1421 |
| GPQA Diamond | 26.3% | — |
| ARC (AI2) Challenge | 78.3% | — |
| BoolQ | 88.6% | — |
| MMLU | 69.9% | — |
| OpenBookQA | 60.2% | — |
| TriviaQA | 87.6% | — |
Multilingual Solar Pro4 leads
Llama 2-70B: 27.9 (#269), Solar Pro4: 49.0 (#134)
| Benchmark | Llama 2-70B | Solar Pro4 |
|---|---|---|
| LMArena Non-English | 1045 | 1362 |
| LMArena Chinese | 994 | 1417 |
| LMArena French | 1089 | 1403 |
| LMArena German | 1041 | 1351 |
| LMArena Japanese | 927 | 1300 |
| LMArena Korean | 964 | 1383 |
| LMArena Russian | 1083 | 1360 |
| LMArena Spanish | 1142 | 1408 |
Instruction Following Solar Pro4 leads
Llama 2-70B: 55.1 (#273), Solar Pro4: 72.8 (#128)
| Benchmark | Llama 2-70B | Solar Pro4 |
|---|---|---|
| LMArena Instruction Following | 1071 | 1376 |
Long Context Solar Pro4 leads
Llama 2-70B: 32.2 (#265), Solar Pro4: 42.0 (#125)
| Benchmark | Llama 2-70B | Solar Pro4 |
|---|---|---|
| LMArena Longer Query | 1062 | 1379 |
Writing & Preference Solar Pro4 leads
Llama 2-70B: 32.5 (#274), Solar Pro4: 56.7 (#133)
| Benchmark | Llama 2-70B | Solar Pro4 |
|---|---|---|
| LMArena Text | 1115 | 1386 |
| LMArena Creative Writing | 1075 | 1313 |
| LMArena Multi-Turn | 1088 | 1391 |
Frequently asked questions
Is Llama 2-70B better than Solar Pro4?
Solar Pro4 is the stronger model overall, scoring 41.5 to 24.1 on the Noometry Index.
Is Llama 2-70B or Solar Pro4 better for coding?
Solar Pro4 scores higher on coding benchmarks: 39.9 versus 30.8 in the Noometry coding category.
How many benchmarks do Llama 2-70B and Solar Pro4 share?
17 benchmarks have published results for both models. Llama 2-70B has 35 scored results on Noometry and Solar Pro4 has 18.