Model comparison
Inkling-Small vs Qwen2.5 72B Instruct
Inkling-Small is the stronger model overall, scoring 46.5 to 31.9 on the Noometry Index.
Last verified . 20 shared benchmarks.
Alibaba (Qwen)Summary
- They share 20 benchmarks with published results for both. Inkling-Small scores higher in 8 categories and Qwen2.5 72B Instruct in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Inkling-Small leads 45.1 to 19.3.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 90% for Inkling-Small and 8.1% for Qwen2.5 72B Instruct.
- Inkling-Small is cheaper at $0.45 / $1.20 per million input/output tokens, against $1.40 / $5.60 for Qwen2.5 72B Instruct.
- Inkling-Small accepts more context: 524K tokens versus 131K.
Side by side
| Inkling-Small | Qwen2.5 72B Instruct | |
|---|---|---|
| Provider | Thinking Machines Lab | Alibaba (Qwen) |
| Noometry Index | 46.5 | 31.9 |
| Released | 2026-07-15 | 2024-09 |
| Weights | Open | Open |
| Context window | 524K | 131K |
| Max output | 1.05M | 8K |
| Input $ / M tokens | $0.45 | $1.40 |
| Output $ / M tokens | $1.20 | $5.60 |
| Results tracked | 33 | 43 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Inkling-Small leads
Inkling-Small: 43.6 (#85), Qwen2.5 72B Instruct: 33.2 (#260)
| Benchmark | Inkling-Small | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Coding | 1451 | 1292 |
| LMArena WebDev | 1409 | — |
| SciCode | 48.7% | — |
| WeirdML | — | 16% |
| BigCodeBench Instruct | — | 45.8% |
| BigCodeBench Complete | — | 55.9% |
Agentic & Tool Use Not comparable
Inkling-Small: —, Qwen2.5 72B Instruct: 22.1 (#133)
| Benchmark | Inkling-Small | Qwen2.5 72B Instruct |
|---|---|---|
| TheAgentCompany | — | 5.7% |
| BALROG | — | 16.2% |
| METR Time Horizons | — | 35.8% |
Reasoning Inkling-Small leads
Inkling-Small: 38.6 (#63), Qwen2.5 72B Instruct: 22.3 (#199)
| Benchmark | Inkling-Small | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Hard Prompts | 1423 | 1271 |
| Epoch Capabilities Index | 150.15 | 129 |
| ARC-AGI-2 | 40.1% | — |
| ARC-AGI-1 | 84% | — |
| CritPt | 8.3% | — |
| Chess Puzzles | 18% | — |
| Mystery Game Puzzles | 6% | — |
| DTBench | — | 62.9% |
| LMCA | — | 13.4% |
| BIG-Bench Hard | — | 79.8% |
| ForecastBench | — | 57.5 |
| HellaSwag | — | 84.8% |
| PIQA | — | 82.6% |
| WinoGrande | — | 82.3% |
Math Inkling-Small leads
Inkling-Small: 45.1 (#77), Qwen2.5 72B Instruct: 19.3 (#287)
| Benchmark | Inkling-Small | Qwen2.5 72B Instruct |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 90% | 8.1% |
| LMArena Math | 1459 | 1283 |
| FrontierMath (Tiers 1-3) | 46.3% | — |
| FrontierMath Tier 4 | 17.1% | — |
| ProofBench | 6% | — |
| Omni-MATH | — | 33% |
| MATH Level 5 | — | 63.2% |
Knowledge Inkling-Small leads
Inkling-Small: 48.2 (#77), Qwen2.5 72B Instruct: 27.0 (#253)
| Benchmark | Inkling-Small | Qwen2.5 72B Instruct |
|---|---|---|
| GPQA Diamond | 88.5% | 49.1% |
| LMArena Expert | 1442 | 1245 |
| SimpleQA Verified | 19.1% | — |
| MMLU-Pro | — | 63.1% |
| Confabulations | — | 19.1% |
| GPQA (HELM) | — | 42.6% |
| ARC (AI2) Challenge | — | 94.5% |
| MMLU | — | 85.3% |
| TriviaQA | — | 71.9% |
Multimodal Not comparable
Inkling-Small: 39.1 (#62), Qwen2.5 72B Instruct: —
| Benchmark | Inkling-Small | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Vision | 1235 | — |
Multilingual Inkling-Small leads
Inkling-Small: 51.7 (#104), Qwen2.5 72B Instruct: 41.0 (#213)
| Benchmark | Inkling-Small | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Non-English | 1402 | 1252 |
| LMArena Chinese | 1465 | 1272 |
| LMArena French | 1436 | 1280 |
| LMArena German | 1405 | 1234 |
| LMArena Japanese | 1405 | 1180 |
| LMArena Korean | 1363 | 1188 |
| LMArena Russian | 1391 | 1264 |
| LMArena Spanish | 1428 | 1256 |
Instruction Following Inkling-Small leads
Inkling-Small: 73.8 (#114), Qwen2.5 72B Instruct: 65.5 (#221)
| Benchmark | Inkling-Small | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Instruction Following | 1399 | 1254 |
| IFEval | — | 80.6% |
Long Context Inkling-Small leads
Inkling-Small: 42.7 (#118), Qwen2.5 72B Instruct: 38.9 (#188)
| Benchmark | Inkling-Small | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Longer Query | 1401 | 1282 |
Writing & Preference Inkling-Small leads
Inkling-Small: 59.6 (#107), Qwen2.5 72B Instruct: 46.7 (#215)
| Benchmark | Inkling-Small | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Text | 1414 | 1269 |
| LMArena Creative Writing | 1331 | 1221 |
| LMArena Multi-Turn | 1418 | 1272 |
| EQ-Bench Creative Writing | 1491 | — |
| WildBench | — | 80.2% |
Frequently asked questions
Is Inkling-Small better than Qwen2.5 72B Instruct?
Inkling-Small is the stronger model overall, scoring 46.5 to 31.9 on the Noometry Index.
Which is cheaper, Inkling-Small or Qwen2.5 72B Instruct?
Inkling-Small is cheaper. It lists at $0.45 per million input tokens and $1.20 per million output tokens; Qwen2.5 72B Instruct lists at $1.40 and $5.60.
Is Inkling-Small or Qwen2.5 72B Instruct better for coding?
Inkling-Small scores higher on coding benchmarks: 43.6 versus 33.2 in the Noometry coding category.
Which has the bigger context window?
Inkling-Small does, with 524K tokens against 131K.
How many benchmarks do Inkling-Small and Qwen2.5 72B Instruct share?
20 benchmarks have published results for both models. Inkling-Small has 33 scored results on Noometry and Qwen2.5 72B Instruct has 43.