Model comparison
Inkling-Small vs Qwen3.7 Max
Qwen3.7 Max is the stronger model overall, scoring 51.4 to 46.5 on the Noometry Index. Inkling-Small costs 5.9× less per token, which makes it the better buy when Qwen3.7 Max's lead doesn't matter for your workload.
Last verified . 24 shared benchmarks.
Alibaba (Qwen)Summary
- They share 24 benchmarks with published results for both. Inkling-Small scores higher in 0 categories and Qwen3.7 Max in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Qwen3.7 Max leads 62.3 to 44.9.
- The biggest single-benchmark swing is SimpleQA Verified: 19.1% for Inkling-Small and 55.8% for Qwen3.7 Max.
- Inkling-Small is cheaper at $0.45 / $1.20 per million input/output tokens, against $2.50 / $7.50 for Qwen3.7 Max.
- Qwen3.7 Max accepts more context: 1M tokens versus 524K.
- Inkling-Small has downloadable open weights; the other is API-only.
Side by side
| Inkling-Small | Qwen3.7 Max | |
|---|---|---|
| Provider | Thinking Machines Lab | Alibaba (Qwen) |
| Noometry Index | 46.5 | 51.4 |
| Released | 2026-07-15 | 2026-05-19 |
| Weights | Open | Proprietary |
| Context window | 524K | 1M |
| Max output | 1.05M | 131K |
| Input $ / M tokens | $0.45 | $2.50 |
| Output $ / M tokens | $1.20 | $7.50 |
| Results tracked | 33 | 33 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen3.7 Max leads
Inkling-Small: 43.3 (#80), Qwen3.7 Max: 50.2 (#43)
| Benchmark | Inkling-Small | Qwen3.7 Max |
|---|---|---|
| LMArena WebDev | 1408 | 1515 |
| SciCode | 48.7% | 48.8% |
| LMArena Coding | 1451 | 1498 |
| SWE-bench Verified | — | 77.3% |
| ALE-Bench | — | 1,189 |
Agentic & Tool Use Not comparable
Inkling-Small: —, Qwen3.7 Max: 21.8 (#129)
| Benchmark | Inkling-Small | Qwen3.7 Max |
|---|---|---|
| GBAEval | — | 0.4% |
Reasoning Qwen3.7 Max leads
Inkling-Small: 39.0 (#53), Qwen3.7 Max: 48.4 (#36)
| Benchmark | Inkling-Small | Qwen3.7 Max |
|---|---|---|
| CritPt | 8.3% | 13.4% |
| Chess Puzzles | 18% | 19% |
| LMArena Hard Prompts | 1424 | 1483 |
| Mystery Game Puzzles | 6% | 32% |
| Epoch Capabilities Index | 150.13 | 153.74 |
| ARC-AGI-2 | 40.1% | — |
| SimpleBench | — | 70.4% |
| NYT Connections (extended) | — | 85.1% |
| ARC-AGI-1 | 84% | — |
| EBR-Bench | — | 9.5% |
| DTBench | — | 92.3% |
| LMCA | — | 44% |
Math Qwen3.7 Max leads
Inkling-Small: 44.9 (#65), Qwen3.7 Max: 62.3 (#26)
| Benchmark | Inkling-Small | Qwen3.7 Max |
|---|---|---|
| FrontierMath (Tiers 1-3) | 46.3% | 64.6% |
| FrontierMath Tier 4 | 17.1% | 34.1% |
| OTIS Mock AIME 2024-2025 | 90% | 95.6% |
| ProofBench | 6% | 26% |
| LMArena Math | 1453 | 1488 |
Knowledge Qwen3.7 Max leads
Inkling-Small: 48.3 (#71), Qwen3.7 Max: 61.7 (#24)
| Benchmark | Inkling-Small | Qwen3.7 Max |
|---|---|---|
| GPQA Diamond | 88.5% | 90.9% |
| SimpleQA Verified | 19.1% | 55.8% |
| LMArena Expert | 1442 | 1488 |
Multimodal Not comparable
Inkling-Small: 38.4 (#59), Qwen3.7 Max: —
| Benchmark | Inkling-Small | Qwen3.7 Max |
|---|---|---|
| LMArena Vision | 1235 | — |
Multilingual Qwen3.7 Max leads
Inkling-Small: 51.7 (#102), Qwen3.7 Max: 57.1 (#13)
| Benchmark | Inkling-Small | Qwen3.7 Max |
|---|---|---|
| LMArena Non-English | 1400 | 1475 |
| LMArena Chinese | 1458 | 1533 |
| LMArena Russian | 1392 | 1482 |
| LMArena French | 1438 | — |
| LMArena German | 1412 | — |
| LMArena Japanese | 1399 | — |
| LMArena Korean | 1355 | — |
| LMArena Spanish | 1426 | — |
Instruction Following Qwen3.7 Max leads
Inkling-Small: 73.9 (#112), Qwen3.7 Max: 76.9 (#34)
| Benchmark | Inkling-Small | Qwen3.7 Max |
|---|---|---|
| LMArena Instruction Following | 1397 | 1460 |
Long Context Qwen3.7 Max leads
Inkling-Small: 42.7 (#113), Qwen3.7 Max: 45.4 (#34)
| Benchmark | Inkling-Small | Qwen3.7 Max |
|---|---|---|
| LMArena Longer Query | 1401 | 1484 |
Writing & Preference Qwen3.7 Max leads
Inkling-Small: 59.8 (#102), Qwen3.7 Max: 65.2 (#52)
| Benchmark | Inkling-Small | Qwen3.7 Max |
|---|---|---|
| LMArena Text | 1413 | 1476 |
| LMArena Creative Writing | 1330 | 1450 |
| LMArena Multi-Turn | 1420 | 1482 |
| EQ-Bench Creative Writing | 1491 | — |
| EQ-Bench 4 | — | 1110 |
Frequently asked questions
Is Inkling-Small better than Qwen3.7 Max?
Qwen3.7 Max is the stronger model overall, scoring 51.4 to 46.5 on the Noometry Index. Inkling-Small costs 5.9× less per token, which makes it the better buy when Qwen3.7 Max's lead doesn't matter for your workload.
Which is cheaper, Inkling-Small or Qwen3.7 Max?
Inkling-Small is cheaper. It lists at $0.45 per million input tokens and $1.20 per million output tokens; Qwen3.7 Max lists at $2.50 and $7.50.
Is Inkling-Small or Qwen3.7 Max better for coding?
Qwen3.7 Max scores higher on coding benchmarks: 50.2 versus 43.3 in the Noometry coding category.
Which has the bigger context window?
Qwen3.7 Max does, with 1M tokens against 524K.
How many benchmarks do Inkling-Small and Qwen3.7 Max share?
24 benchmarks have published results for both models. Inkling-Small has 33 scored results on Noometry and Qwen3.7 Max has 33.