Model comparison
Granite 3.1 8b Instruct vs Qwen3.7 Max
Qwen3.7 Max is the stronger model overall, scoring 51.5 to 32.4 on the Noometry Index.
Last verified . 12 shared benchmarks.
Alibaba (Qwen)Summary
- They share 12 benchmarks with published results for both. Granite 3.1 8b Instruct scores higher in 1 category and Qwen3.7 Max in 8 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Qwen3.7 Max leads 61.6 to 31.1.
- Granite 3.1 8b Instruct has downloadable open weights; the other is API-only.
Side by side
| Granite 3.1 8b Instruct | Qwen3.7 Max | |
|---|---|---|
| Provider | IBM | Alibaba (Qwen) |
| Noometry Index | 32.4 | 51.5 |
| Released | — | 2026-05-19 |
| Weights | Open | Proprietary |
| Context window | — | 1M |
| Max output | — | 131K |
| Input $ / M tokens | — | $2.50 |
| Output $ / M tokens | — | $7.50 |
| Results tracked | 13 | 33 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen3.7 Max leads
Granite 3.1 8b Instruct: 34.5 (#233), Qwen3.7 Max: 50.4 (#45)
| Benchmark | Granite 3.1 8b Instruct | Qwen3.7 Max |
|---|---|---|
| LMArena Coding | 1186 | 1498 |
| SWE-bench Verified | — | 77.3% |
| LMArena WebDev | — | 1515 |
| SciCode | — | 48.8% |
| ALE-Bench | — | 1,189 |
Agentic & Tool Use Granite 3.1 8b Instruct leads
Granite 3.1 8b Instruct: 24.1 (#120), Qwen3.7 Max: 22.1 (#135)
| Benchmark | Granite 3.1 8b Instruct | Qwen3.7 Max |
|---|---|---|
| Berkeley Function Calling Leaderboard | 27.1% | — |
| GBAEval | — | 0.4% |
Reasoning Qwen3.7 Max leads
Granite 3.1 8b Instruct: 22.1 (#207), Qwen3.7 Max: 49.2 (#38)
| Benchmark | Granite 3.1 8b Instruct | Qwen3.7 Max |
|---|---|---|
| LMArena Hard Prompts | 1145 | 1483 |
| SimpleBench | — | 70.4% |
| NYT Connections (extended) | — | 85.1% |
| CritPt | — | 13.4% |
| Chess Puzzles | — | 19% |
| EBR-Bench | — | 9.5% |
| Mystery Game Puzzles | — | 32% |
| DTBench | — | 92.3% |
| LMCA | — | 44% |
| Epoch Capabilities Index | — | 153.68 |
Math Qwen3.7 Max leads
Granite 3.1 8b Instruct: 33.0 (#209), Qwen3.7 Max: 62.4 (#32)
| Benchmark | Granite 3.1 8b Instruct | Qwen3.7 Max |
|---|---|---|
| LMArena Math | 1152 | 1490 |
| FrontierMath (Tiers 1-3) | — | 64.6% |
| FrontierMath Tier 4 | — | 34.1% |
| OTIS Mock AIME 2024-2025 | — | 95.6% |
| ProofBench | — | 26% |
Knowledge Qwen3.7 Max leads
Granite 3.1 8b Instruct: 31.1 (#220), Qwen3.7 Max: 61.6 (#28)
| Benchmark | Granite 3.1 8b Instruct | Qwen3.7 Max |
|---|---|---|
| LMArena Expert | 1142 | 1488 |
| GPQA Diamond | — | 90.9% |
| SimpleQA Verified | — | 55.8% |
Multilingual Qwen3.7 Max leads
Granite 3.1 8b Instruct: 30.9 (#260), Qwen3.7 Max: 56.9 (#15)
| Benchmark | Granite 3.1 8b Instruct | Qwen3.7 Max |
|---|---|---|
| LMArena Non-English | 1099 | 1474 |
| LMArena Chinese | 1145 | 1530 |
| LMArena Russian | 1092 | 1484 |
Instruction Following Qwen3.7 Max leads
Granite 3.1 8b Instruct: 58.6 (#259), Qwen3.7 Max: 76.7 (#38)
| Benchmark | Granite 3.1 8b Instruct | Qwen3.7 Max |
|---|---|---|
| LMArena Instruction Following | 1131 | 1460 |
Long Context Qwen3.7 Max leads
Granite 3.1 8b Instruct: 35.2 (#241), Qwen3.7 Max: 45.4 (#40)
| Benchmark | Granite 3.1 8b Instruct | Qwen3.7 Max |
|---|---|---|
| LMArena Longer Query | 1162 | 1482 |
Writing & Preference Qwen3.7 Max leads
Granite 3.1 8b Instruct: 35.5 (#266), Qwen3.7 Max: 65.0 (#54)
| Benchmark | Granite 3.1 8b Instruct | Qwen3.7 Max |
|---|---|---|
| LMArena Text | 1150 | 1476 |
| LMArena Creative Writing | 1129 | 1449 |
| LMArena Multi-Turn | 1108 | 1481 |
| EQ-Bench 4 | — | 1110 |
Frequently asked questions
Is Granite 3.1 8b Instruct better than Qwen3.7 Max?
Qwen3.7 Max is the stronger model overall, scoring 51.5 to 32.4 on the Noometry Index.
Is Granite 3.1 8b Instruct or Qwen3.7 Max better for coding?
Qwen3.7 Max scores higher on coding benchmarks: 50.4 versus 34.5 in the Noometry coding category.
How many benchmarks do Granite 3.1 8b Instruct and Qwen3.7 Max share?
12 benchmarks have published results for both models. Granite 3.1 8b Instruct has 13 scored results on Noometry and Qwen3.7 Max has 33.