Model comparison
Inkling-Small vs Qwen2.5 32B Instruct
Inkling-Small is the stronger model overall, scoring 46.5 to 30.1 on the Noometry Index.
Last verified . 4 shared benchmarks.
Alibaba (Qwen)Summary
- They share 4 benchmarks with published results for both. Inkling-Small scores higher in 4 categories and Qwen2.5 32B Instruct in 0 categories; 4 gaps are clear of the uncertainty.
- The widest gap is in math, where Inkling-Small leads 45.1 to 16.2.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 90% for Inkling-Small and 7.4% for Qwen2.5 32B Instruct.
- Inkling-Small is cheaper at $0.45 / $1.20 per million input/output tokens, against $0.70 / $2.80 for Qwen2.5 32B Instruct.
- Inkling-Small accepts more context: 524K tokens versus 131K.
Side by side
| Inkling-Small | Qwen2.5 32B Instruct | |
|---|---|---|
| Provider | Thinking Machines Lab | Alibaba (Qwen) |
| Noometry Index | 46.5 | 30.1 |
| Released | 2026-07-15 | 2024-09 |
| Weights | Open | Open |
| Context window | 524K | 131K |
| Max output | 1.05M | 8K |
| Input $ / M tokens | $0.45 | $0.70 |
| Output $ / M tokens | $1.20 | $2.80 |
| Results tracked | 33 | 7 |
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 32B Instruct: 38.7 (#169)
| Benchmark | Inkling-Small | Qwen2.5 32B Instruct |
|---|---|---|
| LMArena WebDev | 1409 | — |
| SciCode | 48.7% | — |
| BigCodeBench Instruct | — | 45% |
| LMArena Coding | 1451 | — |
| BigCodeBench Complete | — | 52.3% |
Reasoning Inkling-Small leads
Inkling-Small: 38.6 (#63), Qwen2.5 32B Instruct: 19.2 (#266)
| Benchmark | Inkling-Small | Qwen2.5 32B Instruct |
|---|---|---|
| Chess Puzzles | 18% | 0% |
| Epoch Capabilities Index | 150.15 | 128.52 |
| ARC-AGI-2 | 40.1% | — |
| ARC-AGI-1 | 84% | — |
| CritPt | 8.3% | — |
| LMArena Hard Prompts | 1423 | — |
| Mystery Game Puzzles | 6% | — |
Math Inkling-Small leads
Inkling-Small: 45.1 (#77), Qwen2.5 32B Instruct: 16.2 (#296)
| Benchmark | Inkling-Small | Qwen2.5 32B Instruct |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 90% | 7.4% |
| FrontierMath (Tiers 1-3) | 46.3% | — |
| FrontierMath Tier 4 | 17.1% | — |
| ProofBench | 6% | — |
| LMArena Math | 1459 | — |
| MATH Level 5 | — | 56.1% |
Knowledge Inkling-Small leads
Inkling-Small: 48.2 (#77), Qwen2.5 32B Instruct: 24.9 (#266)
| Benchmark | Inkling-Small | Qwen2.5 32B Instruct |
|---|---|---|
| GPQA Diamond | 88.5% | 46.1% |
| SimpleQA Verified | 19.1% | — |
| LMArena Expert | 1442 | — |
Multimodal Not comparable
Inkling-Small: 39.1 (#62), Qwen2.5 32B Instruct: —
| Benchmark | Inkling-Small | Qwen2.5 32B Instruct |
|---|---|---|
| LMArena Vision | 1235 | — |
Multilingual Not comparable
Inkling-Small: 51.7 (#104), Qwen2.5 32B Instruct: —
| Benchmark | Inkling-Small | Qwen2.5 32B Instruct |
|---|---|---|
| LMArena Non-English | 1402 | — |
| LMArena Chinese | 1465 | — |
| LMArena French | 1436 | — |
| LMArena German | 1405 | — |
| LMArena Japanese | 1405 | — |
| LMArena Korean | 1363 | — |
| LMArena Russian | 1391 | — |
| LMArena Spanish | 1428 | — |
Instruction Following Not comparable
Inkling-Small: 73.8 (#114), Qwen2.5 32B Instruct: —
| Benchmark | Inkling-Small | Qwen2.5 32B Instruct |
|---|---|---|
| LMArena Instruction Following | 1399 | — |
Long Context Not comparable
Inkling-Small: 42.7 (#118), Qwen2.5 32B Instruct: —
| Benchmark | Inkling-Small | Qwen2.5 32B Instruct |
|---|---|---|
| LMArena Longer Query | 1401 | — |
Writing & Preference Not comparable
Inkling-Small: 59.6 (#107), Qwen2.5 32B Instruct: —
| Benchmark | Inkling-Small | Qwen2.5 32B Instruct |
|---|---|---|
| LMArena Text | 1414 | — |
| LMArena Creative Writing | 1331 | — |
| EQ-Bench Creative Writing | 1491 | — |
| LMArena Multi-Turn | 1418 | — |
Frequently asked questions
Is Inkling-Small better than Qwen2.5 32B Instruct?
Inkling-Small is the stronger model overall, scoring 46.5 to 30.1 on the Noometry Index.
Which is cheaper, Inkling-Small or Qwen2.5 32B Instruct?
Inkling-Small is cheaper. It lists at $0.45 per million input tokens and $1.20 per million output tokens; Qwen2.5 32B Instruct lists at $0.70 and $2.80.
Is Inkling-Small or Qwen2.5 32B Instruct better for coding?
Inkling-Small scores higher on coding benchmarks: 43.6 versus 38.7 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 32B Instruct share?
4 benchmarks have published results for both models. Inkling-Small has 33 scored results on Noometry and Qwen2.5 32B Instruct has 7.