Model comparison
Inkling vs Qwen3-30B-A3B
Inkling is the stronger model overall, scoring 44.1 to 38.9 on the Noometry Index. Qwen3-30B-A3B costs 12× less per token, which makes it the better buy when Inkling's lead doesn't matter for your workload.
Last verified . 26 shared benchmarks.
Alibaba (Qwen)Summary
- They share 26 benchmarks with published results for both. Inkling scores higher in 6 categories and Qwen3-30B-A3B in 3 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Inkling leads 40.4 to 22.2.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 88.9% for Inkling and 70.3% for Qwen3-30B-A3B.
- Qwen3-30B-A3B is cheaper at $0.12 / $0.50 per million input/output tokens, against $1.87 / $4.68 for Inkling.
- Inkling accepts more context: 66K tokens versus 41K.
Side by side
| Inkling | Qwen3-30B-A3B | |
|---|---|---|
| Provider | Thinking Machines Lab | Alibaba (Qwen) |
| Noometry Index | 44.1 | 38.9 |
| Released | 2026-07-15 | 2025-04-28 |
| Weights | Open | Open |
| Context window | 66K | 41K |
| Max output | 66K | 16K |
| Input $ / M tokens | $1.87 | $0.12 |
| Output $ / M tokens | $4.68 | $0.50 |
| Results tracked | 41 | 32 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen3-30B-A3B leads
Inkling: 34.5 (#234), Qwen3-30B-A3B: 37.5 (#194)
| Benchmark | Inkling | Qwen3-30B-A3B |
|---|---|---|
| SciCode | 47% | 33.3% |
| WeirdML | 32.3% | 29.8% |
| LMArena Coding | 1464 | 1416 |
| FrontierCode | 14% | — |
| LMArena WebDev | 1413 | — |
| FrontierSWE | 4.1% | — |
| ALE-Bench | 946 | — |
Agentic & Tool Use Too close to call
Inkling: 29.6 (#85), Qwen3-30B-A3B: 29.8 (#82)
| Benchmark | Inkling | Qwen3-30B-A3B |
|---|---|---|
| APEX-Agents | 33.8% | — |
| Berkeley Function Calling Leaderboard | — | 41.4% |
| τ²-bench Banking | 25% | — |
Reasoning Inkling leads
Inkling: 40.4 (#56), Qwen3-30B-A3B: 22.2 (#204)
| Benchmark | Inkling | Qwen3-30B-A3B |
|---|---|---|
| CritPt | 5.4% | 0.3% |
| Chess Puzzles | 21% | 8% |
| LMArena Hard Prompts | 1451 | 1398 |
| DTBench | 87.5% | 69.3% |
| LMCA | 37.6% | 22.4% |
| Epoch Capabilities Index | 148.54 | 139.63 |
| ARC-AGI-2 | 36.5% | — |
| SimpleBench | 50% | — |
| Kagi LLM Benchmark | — | 54.9% |
| ARC-AGI-1 | 79.5% | — |
Math Qwen3-30B-A3B leads
Inkling: 31.3 (#225), Qwen3-30B-A3B: 37.4 (#157)
| Benchmark | Inkling | Qwen3-30B-A3B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 88.9% | 70.3% |
| LMArena Math | 1479 | 1394 |
| FrontierMath (Tiers 1-3) | 33.3% | — |
| FrontierMath Tier 4 | 4.9% | — |
| MathArena Final-Answer Competitions | — | 47.8% |
| ProofBench | 0% | — |
Knowledge Inkling leads
Inkling: 55.1 (#49), Qwen3-30B-A3B: 41.8 (#105)
| Benchmark | Inkling | Qwen3-30B-A3B |
|---|---|---|
| GPQA Diamond | 88.3% | 70.1% |
| LMArena Expert | 1465 | 1396 |
| SimpleQA Verified | 40.3% | — |
| Confabulations | — | 12.3% |
Multilingual Inkling leads
Inkling: 54.0 (#52), Qwen3-30B-A3B: 49.5 (#132)
| Benchmark | Inkling | Qwen3-30B-A3B |
|---|---|---|
| LMArena Non-English | 1434 | 1372 |
| LMArena Chinese | 1490 | 1433 |
| LMArena French | 1458 | 1418 |
| LMArena German | 1446 | 1380 |
| LMArena Japanese | 1429 | 1337 |
| LMArena Korean | 1404 | 1331 |
| LMArena Russian | 1429 | 1370 |
| LMArena Spanish | 1448 | 1404 |
Instruction Following Inkling leads
Inkling: 75.1 (#71), Qwen3-30B-A3B: 72.0 (#142)
| Benchmark | Inkling | Qwen3-30B-A3B |
|---|---|---|
| LMArena Instruction Following | 1426 | 1363 |
Long Context Inkling leads
Inkling: 43.8 (#86), Qwen3-30B-A3B: 31.0 (#283)
| Benchmark | Inkling | Qwen3-30B-A3B |
|---|---|---|
| LMArena Longer Query | 1434 | 1379 |
| Fiction.LiveBench | — | 40.6% |
Writing & Preference Inkling leads
Inkling: 65.2 (#51), Qwen3-30B-A3B: 55.6 (#143)
| Benchmark | Inkling | Qwen3-30B-A3B |
|---|---|---|
| LMArena Text | 1441 | 1384 |
| LMArena Creative Writing | 1387 | 1317 |
| LMArena Multi-Turn | 1436 | 1378 |
| Short-Story Creative Writing | — | 75.3% |
| EQ-Bench Creative Writing | 1611 | — |
| EQ-Bench 4 | 1226 | — |
Frequently asked questions
Is Inkling better than Qwen3-30B-A3B?
Inkling is the stronger model overall, scoring 44.1 to 38.9 on the Noometry Index. Qwen3-30B-A3B costs 12× less per token, which makes it the better buy when Inkling's lead doesn't matter for your workload.
Which is cheaper, Inkling or Qwen3-30B-A3B?
Qwen3-30B-A3B is cheaper. It lists at $0.12 per million input tokens and $0.50 per million output tokens; Inkling lists at $1.87 and $4.68.
Is Inkling or Qwen3-30B-A3B better for coding?
Qwen3-30B-A3B scores higher on coding benchmarks: 37.5 versus 34.5 in the Noometry coding category.
Which has the bigger context window?
Inkling does, with 66K tokens against 41K.
How many benchmarks do Inkling and Qwen3-30B-A3B share?
26 benchmarks have published results for both models. Inkling has 41 scored results on Noometry and Qwen3-30B-A3B has 32.