Model comparison
Inkling vs Qwen3 235B-A22B
Inkling and Qwen3 235B-A22B score almost the same on the Noometry Index (44.1 vs 43.5), so choose on price, context window or the category you care about most.
Last verified . 31 shared benchmarks.
Alibaba (Qwen)Summary
- They share 31 benchmarks with published results for both. Inkling scores higher in 5 categories and Qwen3 235B-A22B in 4 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Inkling leads 40.4 to 15.7.
- The biggest single-benchmark swing is ARC-AGI-1: 79.5% for Inkling and 11% for Qwen3 235B-A22B.
- Qwen3 235B-A22B is cheaper at $0.70 / $2.80 per million input/output tokens, against $1.87 / $4.68 for Inkling.
- Qwen3 235B-A22B accepts more context: 131K tokens versus 66K.
Side by side
| Inkling | Qwen3 235B-A22B | |
|---|---|---|
| Provider | Thinking Machines Lab | Alibaba (Qwen) |
| Noometry Index | 44.1 | 43.5 |
| Released | 2026-07-15 | 2025-04 |
| Weights | Open | Open |
| Context window | 66K | 131K |
| Max output | 66K | 16K |
| Input $ / M tokens | $1.87 | $0.70 |
| Output $ / M tokens | $4.68 | $2.80 |
| Results tracked | 41 | 49 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen3 235B-A22B leads
Inkling: 34.5 (#234), Qwen3 235B-A22B: 44.3 (#75)
| Benchmark | Inkling | Qwen3 235B-A22B |
|---|---|---|
| SciCode | 47% | 42.4% |
| WeirdML | 32.3% | 41% |
| LMArena Coding | 1464 | 1445 |
| FrontierCode | 14% | — |
| Aider Polyglot | — | 59.6% |
| LMArena WebDev | 1413 | — |
| FrontierSWE | 4.1% | — |
| ALE-Bench | 946 | — |
Agentic & Tool Use Qwen3 235B-A22B leads
Inkling: 29.6 (#85), Qwen3 235B-A22B: 33.9 (#51)
| Benchmark | Inkling | Qwen3 235B-A22B |
|---|---|---|
| APEX-Agents | 33.8% | — |
| Berkeley Function Calling Leaderboard | — | 52.1% |
| τ²-bench Banking | 25% | — |
| Vending-Bench 2 | — | -11.34 |
Reasoning Inkling leads
Inkling: 40.4 (#56), Qwen3 235B-A22B: 15.7 (#311)
| Benchmark | Inkling | Qwen3 235B-A22B |
|---|---|---|
| ARC-AGI-2 | 36.5% | 1.3% |
| SimpleBench | 50% | 31% |
| ARC-AGI-1 | 79.5% | 11% |
| CritPt | 5.4% | 0% |
| Chess Puzzles | 21% | 12% |
| LMArena Hard Prompts | 1451 | 1433 |
| DTBench | 87.5% | 80.3% |
| LMCA | 37.6% | 29.3% |
| Epoch Capabilities Index | 148.54 | 143.85 |
| Kagi LLM Benchmark | — | 69.4% |
| Mystery Game Puzzles | — | 9% |
| ForecastBench | — | 59.7 |
Math Qwen3 235B-A22B leads
Inkling: 31.3 (#225), Qwen3 235B-A22B: 50.4 (#57)
| Benchmark | Inkling | Qwen3 235B-A22B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 88.9% | 86.7% |
| LMArena Math | 1479 | 1432 |
| FrontierMath (Tiers 1-3) | 33.3% | — |
| FrontierMath Tier 4 | 4.9% | — |
| ProofBench | 0% | — |
| Omni-MATH | — | 71.8% |
| MATH Level 5 | — | 68.9% |
| FrontierMath (Feb 2025 set) | — | 8.5% |
| FrontierMath Tier 4 (v1) | — | 0% |
Knowledge Inkling leads
Inkling: 55.1 (#49), Qwen3 235B-A22B: 49.6 (#73)
| Benchmark | Inkling | Qwen3 235B-A22B |
|---|---|---|
| GPQA Diamond | 88.3% | 80.1% |
| SimpleQA Verified | 40.3% | 40.4% |
| LMArena Expert | 1465 | 1463 |
| MMLU-Pro | — | 84.4% |
| Confabulations | — | 15.6% |
| Vectara Hallucination Rate | — | 9.3% |
| GPQA (HELM) | — | 72.7% |
Multilingual Inkling leads
Inkling: 54.0 (#52), Qwen3 235B-A22B: 52.3 (#89)
| Benchmark | Inkling | Qwen3 235B-A22B |
|---|---|---|
| LMArena Non-English | 1434 | 1409 |
| LMArena Chinese | 1490 | 1481 |
| LMArena French | 1458 | 1445 |
| LMArena German | 1446 | 1433 |
| LMArena Japanese | 1429 | 1399 |
| LMArena Korean | 1404 | 1391 |
| LMArena Russian | 1429 | 1411 |
| LMArena Spanish | 1448 | 1430 |
Instruction Following Inkling leads
Inkling: 75.1 (#71), Qwen3 235B-A22B: 72.6 (#136)
| Benchmark | Inkling | Qwen3 235B-A22B |
|---|---|---|
| LMArena Instruction Following | 1426 | 1408 |
| IFEval | — | 83.5% |
Long Context Qwen3 235B-A22B leads
Inkling: 43.8 (#86), Qwen3 235B-A22B: 46.1 (#26)
| Benchmark | Inkling | Qwen3 235B-A22B |
|---|---|---|
| LMArena Longer Query | 1434 | 1426 |
| Fiction.LiveBench | — | 75% |
Writing & Preference Inkling leads
Inkling: 65.2 (#51), Qwen3 235B-A22B: 59.6 (#108)
| Benchmark | Inkling | Qwen3 235B-A22B |
|---|---|---|
| LMArena Text | 1441 | 1419 |
| LMArena Creative Writing | 1387 | 1384 |
| EQ-Bench Creative Writing | 1611 | 1366 |
| LMArena Multi-Turn | 1436 | 1432 |
| Short-Story Creative Writing | — | 83% |
| WildBench | — | 86.6% |
| EQ-Bench 4 | 1226 | — |
Frequently asked questions
Is Inkling better than Qwen3 235B-A22B?
Inkling and Qwen3 235B-A22B score almost the same on the Noometry Index (44.1 vs 43.5), so choose on price, context window or the category you care about most.
Which is cheaper, Inkling or Qwen3 235B-A22B?
Qwen3 235B-A22B is cheaper. It lists at $0.70 per million input tokens and $2.80 per million output tokens; Inkling lists at $1.87 and $4.68.
Is Inkling or Qwen3 235B-A22B better for coding?
Qwen3 235B-A22B scores higher on coding benchmarks: 44.3 versus 34.5 in the Noometry coding category.
Which has the bigger context window?
Qwen3 235B-A22B does, with 131K tokens against 66K.
How many benchmarks do Inkling and Qwen3 235B-A22B share?
31 benchmarks have published results for both models. Inkling has 41 scored results on Noometry and Qwen3 235B-A22B has 49.