Model comparison
Inkling vs Qwen2-72B
Inkling is the stronger model overall, scoring 44.1 to 30.0 on the Noometry Index.
Last verified . 20 shared benchmarks.
Alibaba (Qwen)Summary
- They share 20 benchmarks with published results for both. Inkling scores higher in 9 categories and Qwen2-72B in 0 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Inkling leads 55.1 to 21.2.
- The biggest single-benchmark swing is GPQA Diamond: 88.3% for Inkling and 40.8% for Qwen2-72B.
Side by side
| Inkling | Qwen2-72B | |
|---|---|---|
| Provider | Thinking Machines Lab | Alibaba (Qwen) |
| Noometry Index | 44.1 | 30.0 |
| Released | 2026-07-15 | 2024-06-07 |
| Weights | Open | Open |
| Context window | 66K | — |
| Max output | 66K | — |
| Input $ / M tokens | $1.87 | — |
| Output $ / M tokens | $4.68 | — |
| Results tracked | 41 | 26 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Inkling leads
Inkling: 34.5 (#234), Qwen2-72B: 29.1 (#310)
| Benchmark | Inkling | Qwen2-72B |
|---|---|---|
| WeirdML | 32.3% | 11.3% |
| LMArena Coding | 1464 | 1196 |
| FrontierCode | 14% | — |
| LMArena WebDev | 1413 | — |
| FrontierSWE | 4.1% | — |
| SciCode | 47% | — |
| BigCodeBench Instruct | — | 38.5% |
| BigCodeBench Complete | — | 54% |
| ALE-Bench | 946 | — |
Agentic & Tool Use Inkling leads
Inkling: 29.6 (#85), Qwen2-72B: 17.0 (#146)
| Benchmark | Inkling | Qwen2-72B |
|---|---|---|
| APEX-Agents | 33.8% | — |
| TheAgentCompany | — | 1.1% |
| τ²-bench Banking | 25% | — |
| METR Time Horizons | — | 29.9% |
Reasoning Inkling leads
Inkling: 40.4 (#56), Qwen2-72B: 23.2 (#181)
| Benchmark | Inkling | Qwen2-72B |
|---|---|---|
| LMArena Hard Prompts | 1451 | 1191 |
| Epoch Capabilities Index | 148.54 | 125.28 |
| ARC-AGI-2 | 36.5% | — |
| SimpleBench | 50% | — |
| ARC-AGI-1 | 79.5% | — |
| CritPt | 5.4% | — |
| Chess Puzzles | 21% | — |
| DTBench | 87.5% | — |
| LMCA | 37.6% | — |
Math Inkling leads
Inkling: 31.3 (#225), Qwen2-72B: 30.2 (#236)
| Benchmark | Inkling | Qwen2-72B |
|---|---|---|
| LMArena Math | 1479 | 1235 |
| FrontierMath (Tiers 1-3) | 33.3% | — |
| FrontierMath Tier 4 | 4.9% | — |
| OTIS Mock AIME 2024-2025 | 88.9% | — |
| ProofBench | 0% | — |
| MATH Level 5 | — | 39.1% |
Knowledge Inkling leads
Inkling: 55.1 (#49), Qwen2-72B: 21.2 (#275)
| Benchmark | Inkling | Qwen2-72B |
|---|---|---|
| GPQA Diamond | 88.3% | 40.8% |
| LMArena Expert | 1465 | 1171 |
| SimpleQA Verified | 40.3% | — |
| MMLU | — | 82.4% |
Multilingual Inkling leads
Inkling: 54.0 (#52), Qwen2-72B: 35.9 (#244)
| Benchmark | Inkling | Qwen2-72B |
|---|---|---|
| LMArena Non-English | 1434 | 1176 |
| LMArena Chinese | 1490 | 1240 |
| LMArena French | 1458 | 1170 |
| LMArena German | 1446 | 1151 |
| LMArena Japanese | 1429 | 1111 |
| LMArena Korean | 1404 | 1083 |
| LMArena Russian | 1429 | 1169 |
| LMArena Spanish | 1448 | 1169 |
Instruction Following Inkling leads
Inkling: 75.1 (#71), Qwen2-72B: 61.7 (#241)
| Benchmark | Inkling | Qwen2-72B |
|---|---|---|
| LMArena Instruction Following | 1426 | 1181 |
Long Context Inkling leads
Inkling: 43.8 (#86), Qwen2-72B: 36.1 (#235)
| Benchmark | Inkling | Qwen2-72B |
|---|---|---|
| LMArena Longer Query | 1434 | 1192 |
Writing & Preference Inkling leads
Inkling: 65.2 (#51), Qwen2-72B: 40.8 (#241)
| Benchmark | Inkling | Qwen2-72B |
|---|---|---|
| LMArena Text | 1441 | 1203 |
| LMArena Creative Writing | 1387 | 1181 |
| LMArena Multi-Turn | 1436 | 1196 |
| EQ-Bench Creative Writing | 1611 | — |
| EQ-Bench 4 | 1226 | — |
Frequently asked questions
Is Inkling better than Qwen2-72B?
Inkling is the stronger model overall, scoring 44.1 to 30.0 on the Noometry Index.
Is Inkling or Qwen2-72B better for coding?
Inkling scores higher on coding benchmarks: 34.5 versus 29.1 in the Noometry coding category.
How many benchmarks do Inkling and Qwen2-72B share?
20 benchmarks have published results for both models. Inkling has 41 scored results on Noometry and Qwen2-72B has 26.