Model comparison
Inkling-Small vs Qwen3.5 397B-A17B
Inkling-Small and Qwen3.5 397B-A17B score almost the same on the Noometry Index (46.5 vs 46.0), so choose on price, context window or the category you care about most.
Last verified . 26 shared benchmarks.
Alibaba (Qwen)Summary
- They share 26 benchmarks with published results for both. Inkling-Small scores higher in 2 categories and Qwen3.5 397B-A17B in 7 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Qwen3.5 397B-A17B leads 53.3 to 48.2.
- The biggest single-benchmark swing is FrontierMath (Tiers 1-3): 46.3% for Inkling-Small and 31.2% for Qwen3.5 397B-A17B.
- Inkling-Small is cheaper at $0.45 / $1.20 per million input/output tokens, against $0.60 / $3.60 for Qwen3.5 397B-A17B.
- Inkling-Small accepts more context: 524K tokens versus 262K.
Side by side
| Inkling-Small | Qwen3.5 397B-A17B | |
|---|---|---|
| Provider | Thinking Machines Lab | Alibaba (Qwen) |
| Noometry Index | 46.5 | 46.0 |
| Released | 2026-07-15 | 2026-02-01 |
| Weights | Open | Open |
| Context window | 524K | 262K |
| Max output | 1.05M | 66K |
| Input $ / M tokens | $0.45 | $0.60 |
| Output $ / M tokens | $1.20 | $3.60 |
| Results tracked | 33 | 36 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Inkling-Small leads
Inkling-Small: 43.6 (#85), Qwen3.5 397B-A17B: 42.0 (#114)
| Benchmark | Inkling-Small | Qwen3.5 397B-A17B |
|---|---|---|
| LMArena WebDev | 1409 | 1400 |
| LMArena Coding | 1451 | 1465 |
| SciCode | 48.7% | — |
Agentic & Tool Use Not comparable
Inkling-Small: —, Qwen3.5 397B-A17B: 33.3 (#53)
| Benchmark | Inkling-Small | Qwen3.5 397B-A17B |
|---|---|---|
| APEX-Agents | — | 24.9% |
| τ²-bench Airline | — | 81.5% |
| τ²-bench Banking | — | 9.8% |
| τ²-bench Retail | — | 84.4% |
| τ²-bench Telecom | — | 97.8% |
Reasoning Inkling-Small leads
Inkling-Small: 38.6 (#63), Qwen3.5 397B-A17B: 34.5 (#70)
| Benchmark | Inkling-Small | Qwen3.5 397B-A17B |
|---|---|---|
| Chess Puzzles | 18% | 13% |
| LMArena Hard Prompts | 1423 | 1448 |
| Mystery Game Puzzles | 6% | 18% |
| Epoch Capabilities Index | 150.15 | 146.65 |
| ARC-AGI-2 | 40.1% | — |
| Kagi LLM Benchmark | — | 73.7% |
| NYT Connections (extended) | — | 58.9% |
| ARC-AGI-1 | 84% | — |
| CritPt | 8.3% | — |
| Thematic Generalization | — | 65.1% |
| DTBench | — | 87.5% |
| LMCA | — | 37.9% |
Math Too close to call
Inkling-Small: 45.1 (#77), Qwen3.5 397B-A17B: 46.1 (#73)
| Benchmark | Inkling-Small | Qwen3.5 397B-A17B |
|---|---|---|
| FrontierMath (Tiers 1-3) | 46.3% | 31.2% |
| OTIS Mock AIME 2024-2025 | 90% | 88.9% |
| LMArena Math | 1459 | 1454 |
| FrontierMath Tier 4 | 17.1% | — |
| ProofBench | 6% | — |
Knowledge Qwen3.5 397B-A17B leads
Inkling-Small: 48.2 (#77), Qwen3.5 397B-A17B: 53.3 (#58)
| Benchmark | Inkling-Small | Qwen3.5 397B-A17B |
|---|---|---|
| GPQA Diamond | 88.5% | 86.4% |
| LMArena Expert | 1442 | 1462 |
| SimpleQA Verified | 19.1% | — |
Multimodal Qwen3.5 397B-A17B leads
Inkling-Small: 39.1 (#62), Qwen3.5 397B-A17B: 40.7 (#44)
| Benchmark | Inkling-Small | Qwen3.5 397B-A17B |
|---|---|---|
| LMArena Vision | 1235 | 1263 |
Multilingual Qwen3.5 397B-A17B leads
Inkling-Small: 51.7 (#104), Qwen3.5 397B-A17B: 53.7 (#59)
| Benchmark | Inkling-Small | Qwen3.5 397B-A17B |
|---|---|---|
| LMArena Non-English | 1402 | 1430 |
| LMArena Chinese | 1465 | 1500 |
| LMArena French | 1436 | 1461 |
| LMArena German | 1405 | 1447 |
| LMArena Japanese | 1405 | 1426 |
| LMArena Korean | 1363 | 1384 |
| LMArena Russian | 1391 | 1429 |
| LMArena Spanish | 1428 | 1441 |
Instruction Following Qwen3.5 397B-A17B leads
Inkling-Small: 73.8 (#114), Qwen3.5 397B-A17B: 75.0 (#77)
| Benchmark | Inkling-Small | Qwen3.5 397B-A17B |
|---|---|---|
| LMArena Instruction Following | 1399 | 1424 |
Long Context Qwen3.5 397B-A17B leads
Inkling-Small: 42.7 (#118), Qwen3.5 397B-A17B: 44.1 (#74)
| Benchmark | Inkling-Small | Qwen3.5 397B-A17B |
|---|---|---|
| LMArena Longer Query | 1401 | 1442 |
Writing & Preference Qwen3.5 397B-A17B leads
Inkling-Small: 59.6 (#107), Qwen3.5 397B-A17B: 62.3 (#79)
| Benchmark | Inkling-Small | Qwen3.5 397B-A17B |
|---|---|---|
| LMArena Text | 1414 | 1438 |
| LMArena Creative Writing | 1331 | 1401 |
| EQ-Bench Creative Writing | 1491 | 1478 |
| LMArena Multi-Turn | 1418 | 1446 |
Frequently asked questions
Is Inkling-Small better than Qwen3.5 397B-A17B?
Inkling-Small and Qwen3.5 397B-A17B score almost the same on the Noometry Index (46.5 vs 46.0), so choose on price, context window or the category you care about most.
Which is cheaper, Inkling-Small or Qwen3.5 397B-A17B?
Inkling-Small is cheaper. It lists at $0.45 per million input tokens and $1.20 per million output tokens; Qwen3.5 397B-A17B lists at $0.60 and $3.60.
Is Inkling-Small or Qwen3.5 397B-A17B better for coding?
Inkling-Small scores higher on coding benchmarks: 43.6 versus 42.0 in the Noometry coding category.
Which has the bigger context window?
Inkling-Small does, with 524K tokens against 262K.
How many benchmarks do Inkling-Small and Qwen3.5 397B-A17B share?
26 benchmarks have published results for both models. Inkling-Small has 33 scored results on Noometry and Qwen3.5 397B-A17B has 36.