Model comparison
Molmo 2 8b vs Qwen2.5 72B Instruct
Molmo 2 8b is the stronger model overall, scoring 39.1 to 31.9 on the Noometry Index.
Last verified . 4 shared benchmarks.
Allen Institute for AI (Ai2)
Alibaba (Qwen)Summary
- They share 4 benchmarks with published results for both. Molmo 2 8b scores higher in 4 categories and Qwen2.5 72B Instruct in 0 categories; 4 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Molmo 2 8b leads 25.6 to 22.3.
Side by side
| Molmo 2 8b | Qwen2.5 72B Instruct | |
|---|---|---|
| Provider | Allen Institute for AI (Ai2) | Alibaba (Qwen) |
| Noometry Index | 39.1 | 31.9 |
| Released | — | 2024-09 |
| Weights | Open | Open |
| Context window | — | 131K |
| Max output | — | 8K |
| Input $ / M tokens | — | $1.40 |
| Output $ / M tokens | — | $5.60 |
| Results tracked | 5 | 43 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Not comparable
Molmo 2 8b: —, Qwen2.5 72B Instruct: 33.2 (#260)
| Benchmark | Molmo 2 8b | Qwen2.5 72B Instruct |
|---|---|---|
| WeirdML | — | 16% |
| BigCodeBench Instruct | — | 45.8% |
| LMArena Coding | — | 1292 |
| BigCodeBench Complete | — | 55.9% |
Agentic & Tool Use Not comparable
Molmo 2 8b: —, Qwen2.5 72B Instruct: 22.1 (#133)
| Benchmark | Molmo 2 8b | Qwen2.5 72B Instruct |
|---|---|---|
| TheAgentCompany | — | 5.7% |
| BALROG | — | 16.2% |
| METR Time Horizons | — | 35.8% |
Reasoning Molmo 2 8b leads
Molmo 2 8b: 25.6 (#146), Qwen2.5 72B Instruct: 22.3 (#199)
| Benchmark | Molmo 2 8b | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Hard Prompts | 1287 | 1271 |
| DTBench | — | 62.9% |
| LMCA | — | 13.4% |
| BIG-Bench Hard | — | 79.8% |
| Epoch Capabilities Index | — | 129 |
| ForecastBench | — | 57.5 |
| HellaSwag | — | 84.8% |
| PIQA | — | 82.6% |
| WinoGrande | — | 82.3% |
Math Not comparable
Molmo 2 8b: —, Qwen2.5 72B Instruct: 19.3 (#287)
| Benchmark | Molmo 2 8b | Qwen2.5 72B Instruct |
|---|---|---|
| OTIS Mock AIME 2024-2025 | — | 8.1% |
| Omni-MATH | — | 33% |
| LMArena Math | — | 1283 |
| MATH Level 5 | — | 63.2% |
Knowledge Not comparable
Molmo 2 8b: —, Qwen2.5 72B Instruct: 27.0 (#253)
| Benchmark | Molmo 2 8b | Qwen2.5 72B Instruct |
|---|---|---|
| GPQA Diamond | — | 49.1% |
| MMLU-Pro | — | 63.1% |
| Confabulations | — | 19.1% |
| GPQA (HELM) | — | 42.6% |
| LMArena Expert | — | 1245 |
| ARC (AI2) Challenge | — | 94.5% |
| MMLU | — | 85.3% |
| TriviaQA | — | 71.9% |
Multimodal Not comparable
Molmo 2 8b: 30.2 (#112), Qwen2.5 72B Instruct: —
| Benchmark | Molmo 2 8b | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Vision | 1081 | — |
Multilingual Molmo 2 8b leads
Molmo 2 8b: 42.7 (#190), Qwen2.5 72B Instruct: 41.0 (#213)
| Benchmark | Molmo 2 8b | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Non-English | 1276 | 1252 |
| LMArena Chinese | — | 1272 |
| LMArena French | — | 1280 |
| LMArena German | — | 1234 |
| LMArena Japanese | — | 1180 |
| LMArena Korean | — | 1188 |
| LMArena Russian | — | 1264 |
| LMArena Spanish | — | 1256 |
Instruction Following Molmo 2 8b leads
Molmo 2 8b: 67.0 (#201), Qwen2.5 72B Instruct: 65.5 (#221)
| Benchmark | Molmo 2 8b | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Instruction Following | 1270 | 1254 |
| IFEval | — | 80.6% |
Long Context Not comparable
Molmo 2 8b: —, Qwen2.5 72B Instruct: 38.9 (#188)
| Benchmark | Molmo 2 8b | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Longer Query | — | 1282 |
Writing & Preference Molmo 2 8b leads
Molmo 2 8b: 49.4 (#191), Qwen2.5 72B Instruct: 46.7 (#215)
| Benchmark | Molmo 2 8b | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Text | 1288 | 1269 |
| LMArena Creative Writing | — | 1221 |
| WildBench | — | 80.2% |
| LMArena Multi-Turn | — | 1272 |
Frequently asked questions
Is Molmo 2 8b better than Qwen2.5 72B Instruct?
Molmo 2 8b is the stronger model overall, scoring 39.1 to 31.9 on the Noometry Index.
How many benchmarks do Molmo 2 8b and Qwen2.5 72B Instruct share?
4 benchmarks have published results for both models. Molmo 2 8b has 5 scored results on Noometry and Qwen2.5 72B Instruct has 43.