Model comparison
Llama 2-70B vs Olmo 3 32b Think
Olmo 3 32b Think is the stronger model overall, scoring 38.7 to 24.4 on the Noometry Index.
Last verified . 14 shared benchmarks.
Meta
Allen Institute for AI (Ai2)Summary
- They share 14 benchmarks with published results for both. Llama 2-70B scores higher in 0 categories and Olmo 3 32b Think in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Olmo 3 32b Think leads 36.5 to 8.1.
Side by side
| Llama 2-70B | Olmo 3 32b Think | |
|---|---|---|
| Provider | Meta | Allen Institute for AI (Ai2) |
| Noometry Index | 24.4 | 38.7 |
| Released | 2023-07-18 | — |
| Weights | Open | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 35 | 14 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Olmo 3 32b Think leads
Llama 2-70B: 31.4 (#286), Olmo 3 32b Think: 38.6 (#172)
| Benchmark | Llama 2-70B | Olmo 3 32b Think |
|---|---|---|
| LMArena Coding | 1079 | 1319 |
Reasoning Olmo 3 32b Think leads
Llama 2-70B: 14.4 (#325), Olmo 3 32b Think: 25.9 (#140)
| Benchmark | Llama 2-70B | Olmo 3 32b Think |
|---|---|---|
| LMArena Hard Prompts | 1073 | 1302 |
| DTBench | 41.6% | — |
| BIG-Bench Hard | 64.9% | — |
| CommonsenseQA 2.0 | 50% | — |
| Epoch Capabilities Index | 113.79 | — |
| ForecastBench | 51.4 | — |
| HellaSwag | 85.3% | — |
| LAMBADA | 78.9% | — |
| PIQA | 82.8% | — |
| WinoGrande | 80.2% | — |
Math Olmo 3 32b Think leads
Llama 2-70B: 8.1 (#326), Olmo 3 32b Think: 36.5 (#165)
| Benchmark | Llama 2-70B | Olmo 3 32b Think |
|---|---|---|
| LMArena Math | 1091 | 1316 |
| OTIS Mock AIME 2024-2025 | 0% | — |
| MATH Level 5 | 3.3% | — |
| GSM8K | 69.6% | — |
Knowledge Olmo 3 32b Think leads
Llama 2-70B: 7.4 (#310), Olmo 3 32b Think: 35.0 (#190)
| Benchmark | Llama 2-70B | Olmo 3 32b Think |
|---|---|---|
| LMArena Expert | 1039 | 1273 |
| GPQA Diamond | 26.3% | — |
| ARC (AI2) Challenge | 78.3% | — |
| BoolQ | 88.6% | — |
| MMLU | 69.9% | — |
| OpenBookQA | 60.2% | — |
| TriviaQA | 87.6% | — |
Multilingual Olmo 3 32b Think leads
Llama 2-70B: 27.7 (#274), Olmo 3 32b Think: 41.2 (#210)
| Benchmark | Llama 2-70B | Olmo 3 32b Think |
|---|---|---|
| LMArena Non-English | 1045 | 1255 |
| LMArena Chinese | 995 | 1300 |
| LMArena French | 1090 | 1291 |
| LMArena German | 1041 | 1290 |
| LMArena Russian | 1083 | 1254 |
| LMArena Japanese | 927 | — |
| LMArena Korean | 964 | — |
| LMArena Spanish | 1143 | — |
Instruction Following Olmo 3 32b Think leads
Llama 2-70B: 54.9 (#278), Olmo 3 32b Think: 67.2 (#198)
| Benchmark | Llama 2-70B | Olmo 3 32b Think |
|---|---|---|
| LMArena Instruction Following | 1071 | 1275 |
Long Context Olmo 3 32b Think leads
Llama 2-70B: 32.3 (#270), Olmo 3 32b Think: 39.4 (#182)
| Benchmark | Llama 2-70B | Olmo 3 32b Think |
|---|---|---|
| LMArena Longer Query | 1062 | 1296 |
Writing & Preference Olmo 3 32b Think leads
Llama 2-70B: 32.3 (#279), Olmo 3 32b Think: 49.1 (#193)
| Benchmark | Llama 2-70B | Olmo 3 32b Think |
|---|---|---|
| LMArena Text | 1115 | 1300 |
| LMArena Creative Writing | 1075 | 1256 |
| LMArena Multi-Turn | 1088 | 1290 |
Frequently asked questions
Is Llama 2-70B better than Olmo 3 32b Think?
Olmo 3 32b Think is the stronger model overall, scoring 38.7 to 24.4 on the Noometry Index.
Is Llama 2-70B or Olmo 3 32b Think better for coding?
Olmo 3 32b Think scores higher on coding benchmarks: 38.6 versus 31.4 in the Noometry coding category.
How many benchmarks do Llama 2-70B and Olmo 3 32b Think share?
14 benchmarks have published results for both models. Llama 2-70B has 35 scored results on Noometry and Olmo 3 32b Think has 14.