Model comparison
Granite 4.1 8b vs Qwen2.5 72B Instruct
Granite 4.1 8b is the stronger model overall, scoring 36.8 to 31.5 on the Noometry Index.
Last verified . 12 shared benchmarks.
Alibaba (Qwen)Summary
- They share 12 benchmarks with published results for both. Granite 4.1 8b scores higher in 6 categories and Qwen2.5 72B Instruct in 2 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in math, where Granite 4.1 8b leads 33.8 to 18.1.
Side by side
| Granite 4.1 8b | Qwen2.5 72B Instruct | |
|---|---|---|
| Provider | IBM | Alibaba (Qwen) |
| Noometry Index | 36.8 | 31.5 |
| Released | — | 2024-09 |
| Weights | Open | Open |
| Context window | — | 131K |
| Max output | — | 8K |
| Input $ / M tokens | — | $1.40 |
| Output $ / M tokens | — | $5.60 |
| Results tracked | 13 | 43 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen2.5 72B Instruct leads
Granite 4.1 8b: 29.9 (#291), Qwen2.5 72B Instruct: 32.8 (#256)
| Benchmark | Granite 4.1 8b | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Coding | 1313 | 1293 |
| LMArena WebDev | 1190 | — |
| WeirdML | — | 16% |
| BigCodeBench Instruct | — | 45.8% |
| BigCodeBench Complete | — | 55.9% |
Agentic & Tool Use Not comparable
Granite 4.1 8b: —, Qwen2.5 72B Instruct: 22.0 (#124)
| Benchmark | Granite 4.1 8b | Qwen2.5 72B Instruct |
|---|---|---|
| TheAgentCompany | — | 5.7% |
| BALROG | — | 16.2% |
| METR Time Horizons | — | 35.8% |
Reasoning Granite 4.1 8b leads
Granite 4.1 8b: 24.6 (#142), Qwen2.5 72B Instruct: 21.4 (#197)
| Benchmark | Granite 4.1 8b | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Hard Prompts | 1294 | 1271 |
| DTBench | — | 62.9% |
| LMCA | — | 13.4% |
| BIG-Bench Hard | — | 79.8% |
| Epoch Capabilities Index | — | 129.01 |
| ForecastBench | — | 57.5 |
| HellaSwag | — | 84.8% |
| PIQA | — | 82.6% |
| WinoGrande | — | 82.3% |
Math Granite 4.1 8b leads
Granite 4.1 8b: 33.8 (#167), Qwen2.5 72B Instruct: 18.1 (#281)
| Benchmark | Granite 4.1 8b | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Math | 1311 | 1283 |
| OTIS Mock AIME 2024-2025 | — | 8.1% |
| Omni-MATH | — | 33% |
| MATH Level 5 | — | 63.2% |
Knowledge Granite 4.1 8b leads
Granite 4.1 8b: 35.5 (#171), Qwen2.5 72B Instruct: 26.9 (#247)
| Benchmark | Granite 4.1 8b | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Expert | 1310 | 1245 |
| GPQA Diamond | — | 49.1% |
| MMLU-Pro | — | 63.1% |
| Confabulations | — | 19.1% |
| GPQA (HELM) | — | 42.6% |
| ARC (AI2) Challenge | — | 94.5% |
| MMLU | — | 85.3% |
| TriviaQA | — | 71.9% |
Multilingual Too close to call
Granite 4.1 8b: 41.8 (#199), Qwen2.5 72B Instruct: 41.2 (#208)
| Benchmark | Granite 4.1 8b | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Non-English | 1261 | 1252 |
| LMArena Chinese | 1335 | 1272 |
| LMArena Russian | 1238 | 1264 |
| LMArena French | — | 1279 |
| LMArena German | — | 1234 |
| LMArena Japanese | — | 1180 |
| LMArena Korean | — | 1188 |
| LMArena Spanish | — | 1255 |
Instruction Following Granite 4.1 8b leads
Granite 4.1 8b: 67.0 (#198), Qwen2.5 72B Instruct: 65.7 (#216)
| Benchmark | Granite 4.1 8b | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Instruction Following | 1268 | 1254 |
| IFEval | — | 80.6% |
Long Context Too close to call
Granite 4.1 8b: 38.7 (#188), Qwen2.5 72B Instruct: 38.9 (#183)
| Benchmark | Granite 4.1 8b | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Longer Query | 1275 | 1282 |
Writing & Preference Granite 4.1 8b leads
Granite 4.1 8b: 48.3 (#199), Qwen2.5 72B Instruct: 46.9 (#210)
| Benchmark | Granite 4.1 8b | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Text | 1291 | 1269 |
| LMArena Creative Writing | 1249 | 1221 |
| LMArena Multi-Turn | 1269 | 1272 |
| WildBench | — | 80.2% |
Frequently asked questions
Is Granite 4.1 8b better than Qwen2.5 72B Instruct?
Granite 4.1 8b is the stronger model overall, scoring 36.8 to 31.5 on the Noometry Index.
Is Granite 4.1 8b or Qwen2.5 72B Instruct better for coding?
Qwen2.5 72B Instruct scores higher on coding benchmarks: 32.8 versus 29.9 in the Noometry coding category.
How many benchmarks do Granite 4.1 8b and Qwen2.5 72B Instruct share?
12 benchmarks have published results for both models. Granite 4.1 8b has 13 scored results on Noometry and Qwen2.5 72B Instruct has 43.