Model comparison
Inkling-Small vs Qwen3 Max
Inkling-Small is the stronger model overall, scoring 46.5 to 43.5 on the Noometry Index.
Last verified . 24 shared benchmarks.
Alibaba (Qwen)Summary
- They share 24 benchmarks with published results for both. Inkling-Small scores higher in 5 categories and Qwen3 Max in 3 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Inkling-Small leads 39.0 to 22.0.
- The biggest single-benchmark swing is SimpleQA Verified: 19.1% for Inkling-Small and 48.7% for Qwen3 Max.
- Inkling-Small is cheaper at $0.45 / $1.20 per million input/output tokens, against $1.20 / $6 for Qwen3 Max.
- Inkling-Small accepts more context: 524K tokens versus 262K.
- Inkling-Small has downloadable open weights; the other is API-only.
Side by side
| Inkling-Small | Qwen3 Max | |
|---|---|---|
| Provider | Thinking Machines Lab | Alibaba (Qwen) |
| Noometry Index | 46.5 | 43.5 |
| Released | 2026-07-15 | 2025-09-23 |
| Weights | Open | Proprietary |
| Context window | 524K | 262K |
| Max output | 1.05M | 66K |
| Input $ / M tokens | $0.45 | $1.20 |
| Output $ / M tokens | $1.20 | $6 |
| Results tracked | 33 | 33 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
Inkling-Small: 43.3 (#80), Qwen3 Max: 42.4 (#93)
| Benchmark | Inkling-Small | Qwen3 Max |
|---|---|---|
| LMArena Coding | 1451 | 1456 |
| LMArena WebDev | 1408 | — |
| SciCode | 48.7% | — |
| ALE-Bench | — | 370.45 |
Agentic & Tool Use Not comparable
Inkling-Small: —, Qwen3 Max: —
| Benchmark | Inkling-Small | Qwen3 Max |
|---|---|---|
| Vending-Bench 2 | — | 71.56 |
Reasoning Inkling-Small leads
Inkling-Small: 39.0 (#53), Qwen3 Max: 22.0 (#186)
| Benchmark | Inkling-Small | Qwen3 Max |
|---|---|---|
| Chess Puzzles | 18% | 4% |
| LMArena Hard Prompts | 1424 | 1448 |
| Mystery Game Puzzles | 6% | 5% |
| Epoch Capabilities Index | 150.13 | 142.41 |
| ARC-AGI-2 | 40.1% | — |
| Kagi LLM Benchmark | — | 72.5% |
| NYT Connections (extended) | — | 30.1% |
| ARC-AGI-1 | 84% | — |
| CritPt | 8.3% | — |
| DTBench | — | 82.1% |
| LMCA | — | 28.3% |
Math Inkling-Small leads
Inkling-Small: 44.9 (#65), Qwen3 Max: 37.6 (#94)
| Benchmark | Inkling-Small | Qwen3 Max |
|---|---|---|
| FrontierMath (Tiers 1-3) | 46.3% | 18.9% |
| OTIS Mock AIME 2024-2025 | 90% | 73.3% |
| LMArena Math | 1453 | 1447 |
| FrontierMath Tier 4 | 17.1% | — |
| ProofBench | 6% | — |
| MATH Level 5 | — | 97.1% |
Knowledge Too close to call
Inkling-Small: 48.3 (#71), Qwen3 Max: 48.1 (#72)
| Benchmark | Inkling-Small | Qwen3 Max |
|---|---|---|
| GPQA Diamond | 88.5% | 72.6% |
| SimpleQA Verified | 19.1% | 48.7% |
| LMArena Expert | 1442 | 1457 |
Multimodal Not comparable
Inkling-Small: 38.4 (#59), Qwen3 Max: —
| Benchmark | Inkling-Small | Qwen3 Max |
|---|---|---|
| LMArena Vision | 1235 | — |
Multilingual Qwen3 Max leads
Inkling-Small: 51.7 (#102), Qwen3 Max: 53.9 (#58)
| Benchmark | Inkling-Small | Qwen3 Max |
|---|---|---|
| LMArena Non-English | 1400 | 1429 |
| LMArena Chinese | 1458 | 1479 |
| LMArena French | 1438 | 1450 |
| LMArena German | 1412 | 1463 |
| LMArena Japanese | 1399 | 1399 |
| LMArena Korean | 1355 | 1397 |
| LMArena Russian | 1392 | 1428 |
| LMArena Spanish | 1426 | 1463 |
Instruction Following Qwen3 Max leads
Inkling-Small: 73.9 (#112), Qwen3 Max: 75.0 (#81)
| Benchmark | Inkling-Small | Qwen3 Max |
|---|---|---|
| LMArena Instruction Following | 1397 | 1420 |
Long Context Inkling-Small leads
Inkling-Small: 42.7 (#113), Qwen3 Max: 41.6 (#129)
| Benchmark | Inkling-Small | Qwen3 Max |
|---|---|---|
| LMArena Longer Query | 1401 | 1439 |
| Fiction.LiveBench | — | 66.7% |
| CL-bench | — | 14.5% |
Writing & Preference Qwen3 Max leads
Inkling-Small: 59.8 (#102), Qwen3 Max: 62.6 (#72)
| Benchmark | Inkling-Small | Qwen3 Max |
|---|---|---|
| LMArena Text | 1413 | 1439 |
| LMArena Creative Writing | 1330 | 1401 |
| LMArena Multi-Turn | 1420 | 1446 |
| EQ-Bench Creative Writing | 1491 | — |
Frequently asked questions
Is Inkling-Small better than Qwen3 Max?
Inkling-Small is the stronger model overall, scoring 46.5 to 43.5 on the Noometry Index.
Which is cheaper, Inkling-Small or Qwen3 Max?
Inkling-Small is cheaper. It lists at $0.45 per million input tokens and $1.20 per million output tokens; Qwen3 Max lists at $1.20 and $6.
Is Inkling-Small or Qwen3 Max better for coding?
They score almost the same on coding (43.3 vs 42.4); test both on your own repository before choosing.
Which has the bigger context window?
Inkling-Small does, with 524K tokens against 262K.
How many benchmarks do Inkling-Small and Qwen3 Max share?
24 benchmarks have published results for both models. Inkling-Small has 33 scored results on Noometry and Qwen3 Max has 33.