Model comparison
Granite 3.1 2b Instruct vs Inkling
Inkling is the stronger model overall, scoring 44.1 to 33.2 on the Noometry Index.
Last verified . 12 shared benchmarks.
Summary
- They share 12 benchmarks with published results for both. Granite 3.1 2b Instruct scores higher in 1 category and Inkling in 7 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Inkling leads 65.2 to 34.1.
Side by side
| Granite 3.1 2b Instruct | Inkling | |
|---|---|---|
| Provider | IBM | Thinking Machines Lab |
| Noometry Index | 33.2 | 44.1 |
| Released | — | 2026-07-15 |
| Weights | Open | Open |
| Context window | — | 66K |
| Max output | — | 66K |
| Input $ / M tokens | — | $1.87 |
| Output $ / M tokens | — | $4.68 |
| Results tracked | 12 | 41 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Inkling leads
Granite 3.1 2b Instruct: 33.4 (#257), Inkling: 34.5 (#234)
| Benchmark | Granite 3.1 2b Instruct | Inkling |
|---|---|---|
| LMArena Coding | 1149 | 1464 |
| FrontierCode | — | 14% |
| LMArena WebDev | — | 1413 |
| FrontierSWE | — | 4.1% |
| SciCode | — | 47% |
| WeirdML | — | 32.3% |
| ALE-Bench | — | 946 |
Agentic & Tool Use Not comparable
Granite 3.1 2b Instruct: —, Inkling: 29.6 (#85)
| Benchmark | Granite 3.1 2b Instruct | Inkling |
|---|---|---|
| APEX-Agents | — | 33.8% |
| τ²-bench Banking | — | 25% |
Reasoning Inkling leads
Granite 3.1 2b Instruct: 22.0 (#209), Inkling: 40.4 (#56)
| Benchmark | Granite 3.1 2b Instruct | Inkling |
|---|---|---|
| LMArena Hard Prompts | 1138 | 1451 |
| ARC-AGI-2 | — | 36.5% |
| SimpleBench | — | 50% |
| ARC-AGI-1 | — | 79.5% |
| CritPt | — | 5.4% |
| Chess Puzzles | — | 21% |
| DTBench | — | 87.5% |
| LMCA | — | 37.6% |
| Epoch Capabilities Index | — | 148.54 |
Math Granite 3.1 2b Instruct leads
Granite 3.1 2b Instruct: 33.1 (#206), Inkling: 31.3 (#225)
| Benchmark | Granite 3.1 2b Instruct | Inkling |
|---|---|---|
| LMArena Math | 1159 | 1479 |
| FrontierMath (Tiers 1-3) | — | 33.3% |
| FrontierMath Tier 4 | — | 4.9% |
| OTIS Mock AIME 2024-2025 | — | 88.9% |
| ProofBench | — | 0% |
Knowledge Inkling leads
Granite 3.1 2b Instruct: 30.8 (#224), Inkling: 55.1 (#49)
| Benchmark | Granite 3.1 2b Instruct | Inkling |
|---|---|---|
| LMArena Expert | 1131 | 1465 |
| GPQA Diamond | — | 88.3% |
| SimpleQA Verified | — | 40.3% |
Multilingual Inkling leads
Granite 3.1 2b Instruct: 29.1 (#269), Inkling: 54.0 (#52)
| Benchmark | Granite 3.1 2b Instruct | Inkling |
|---|---|---|
| LMArena Non-English | 1068 | 1434 |
| LMArena Chinese | 1139 | 1490 |
| LMArena Russian | 1063 | 1429 |
| LMArena French | — | 1458 |
| LMArena German | — | 1446 |
| LMArena Japanese | — | 1429 |
| LMArena Korean | — | 1404 |
| LMArena Spanish | — | 1448 |
Instruction Following Inkling leads
Granite 3.1 2b Instruct: 57.7 (#264), Inkling: 75.1 (#71)
| Benchmark | Granite 3.1 2b Instruct | Inkling |
|---|---|---|
| LMArena Instruction Following | 1116 | 1426 |
Long Context Inkling leads
Granite 3.1 2b Instruct: 35.0 (#244), Inkling: 43.8 (#86)
| Benchmark | Granite 3.1 2b Instruct | Inkling |
|---|---|---|
| LMArena Longer Query | 1155 | 1434 |
Writing & Preference Inkling leads
Granite 3.1 2b Instruct: 34.1 (#274), Inkling: 65.2 (#51)
| Benchmark | Granite 3.1 2b Instruct | Inkling |
|---|---|---|
| LMArena Text | 1127 | 1441 |
| LMArena Creative Writing | 1116 | 1387 |
| LMArena Multi-Turn | 1099 | 1436 |
| EQ-Bench Creative Writing | — | 1611 |
| EQ-Bench 4 | — | 1226 |
Frequently asked questions
Is Granite 3.1 2b Instruct better than Inkling?
Inkling is the stronger model overall, scoring 44.1 to 33.2 on the Noometry Index.
Is Granite 3.1 2b Instruct or Inkling better for coding?
Inkling scores higher on coding benchmarks: 34.5 versus 33.4 in the Noometry coding category.
How many benchmarks do Granite 3.1 2b Instruct and Inkling share?
12 benchmarks have published results for both models. Granite 3.1 2b Instruct has 12 scored results on Noometry and Inkling has 41.