Model comparison
Inkling vs Qwen3-Next 80B-A3B Instruct
Inkling is the stronger model overall, scoring 44.1 to 43.0 on the Noometry Index. Qwen3-Next 80B-A3B Instruct costs 2.9× less per token, which makes it the better buy when Inkling's lead doesn't matter for your workload.
Last verified . 17 shared benchmarks.
Alibaba (Qwen)Summary
- They share 17 benchmarks with published results for both. Inkling scores higher in 6 categories and Qwen3-Next 80B-A3B Instruct in 2 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Inkling leads 55.1 to 41.8.
- Qwen3-Next 80B-A3B Instruct is cheaper at $0.50 / $2 per million input/output tokens, against $1.87 / $4.68 for Inkling.
- Qwen3-Next 80B-A3B Instruct accepts more context: 131K tokens versus 66K.
Side by side
| Inkling | Qwen3-Next 80B-A3B Instruct | |
|---|---|---|
| Provider | Thinking Machines Lab | Alibaba (Qwen) |
| Noometry Index | 44.1 | 43.0 |
| Released | 2026-07-15 | 2025-09 |
| Weights | Open | Open |
| Context window | 66K | 131K |
| Max output | 66K | 33K |
| Input $ / M tokens | $1.87 | $0.50 |
| Output $ / M tokens | $4.68 | $2 |
| Results tracked | 41 | 25 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen3-Next 80B-A3B Instruct leads
Inkling: 34.5 (#234), Qwen3-Next 80B-A3B Instruct: 42.5 (#98)
| Benchmark | Inkling | Qwen3-Next 80B-A3B Instruct |
|---|---|---|
| LMArena Coding | 1464 | 1440 |
| FrontierCode | 14% | — |
| LMArena WebDev | 1413 | — |
| FrontierSWE | 4.1% | — |
| SciCode | 47% | — |
| WeirdML | 32.3% | — |
| ALE-Bench | 946 | — |
Agentic & Tool Use Not comparable
Inkling: 29.6 (#85), Qwen3-Next 80B-A3B Instruct: —
| Benchmark | Inkling | Qwen3-Next 80B-A3B Instruct |
|---|---|---|
| APEX-Agents | 33.8% | — |
| τ²-bench Banking | 25% | — |
Reasoning Inkling leads
Inkling: 40.4 (#56), Qwen3-Next 80B-A3B Instruct: 31.1 (#81)
| Benchmark | Inkling | Qwen3-Next 80B-A3B Instruct |
|---|---|---|
| LMArena Hard Prompts | 1451 | 1428 |
| ARC-AGI-2 | 36.5% | — |
| SimpleBench | 50% | — |
| Kagi LLM Benchmark | — | 66.7% |
| ARC-AGI-1 | 79.5% | — |
| CritPt | 5.4% | — |
| Chess Puzzles | 21% | — |
| DTBench | 87.5% | — |
| LMCA | 37.6% | — |
| Epoch Capabilities Index | 148.54 | — |
Math Qwen3-Next 80B-A3B Instruct leads
Inkling: 31.3 (#225), Qwen3-Next 80B-A3B Instruct: 38.8 (#126)
| Benchmark | Inkling | Qwen3-Next 80B-A3B Instruct |
|---|---|---|
| LMArena Math | 1479 | 1440 |
| FrontierMath (Tiers 1-3) | 33.3% | — |
| FrontierMath Tier 4 | 4.9% | — |
| OTIS Mock AIME 2024-2025 | 88.9% | — |
| ProofBench | 0% | — |
| Omni-MATH | — | 46.7% |
Knowledge Inkling leads
Inkling: 55.1 (#49), Qwen3-Next 80B-A3B Instruct: 41.8 (#106)
| Benchmark | Inkling | Qwen3-Next 80B-A3B Instruct |
|---|---|---|
| LMArena Expert | 1465 | 1417 |
| GPQA Diamond | 88.3% | — |
| SimpleQA Verified | 40.3% | — |
| MMLU-Pro | — | 78.6% |
| Vectara Hallucination Rate | — | 9.3% |
| GPQA (HELM) | — | 63% |
Multilingual Inkling leads
Inkling: 54.0 (#52), Qwen3-Next 80B-A3B Instruct: 52.1 (#93)
| Benchmark | Inkling | Qwen3-Next 80B-A3B Instruct |
|---|---|---|
| LMArena Non-English | 1434 | 1407 |
| LMArena Chinese | 1490 | 1460 |
| LMArena French | 1458 | 1413 |
| LMArena German | 1446 | 1417 |
| LMArena Japanese | 1429 | 1395 |
| LMArena Korean | 1404 | 1364 |
| LMArena Russian | 1429 | 1404 |
| LMArena Spanish | 1448 | 1435 |
Instruction Following Inkling leads
Inkling: 75.1 (#71), Qwen3-Next 80B-A3B Instruct: 70.8 (#159)
| Benchmark | Inkling | Qwen3-Next 80B-A3B Instruct |
|---|---|---|
| LMArena Instruction Following | 1426 | 1389 |
| IFEval | — | 81% |
Long Context Inkling leads
Inkling: 43.8 (#86), Qwen3-Next 80B-A3B Instruct: 37.0 (#223)
| Benchmark | Inkling | Qwen3-Next 80B-A3B Instruct |
|---|---|---|
| LMArena Longer Query | 1434 | 1403 |
| Fiction.LiveBench | — | 55.6% |
Writing & Preference Inkling leads
Inkling: 65.2 (#51), Qwen3-Next 80B-A3B Instruct: 58.0 (#121)
| Benchmark | Inkling | Qwen3-Next 80B-A3B Instruct |
|---|---|---|
| LMArena Text | 1441 | 1417 |
| LMArena Creative Writing | 1387 | 1334 |
| LMArena Multi-Turn | 1436 | 1416 |
| EQ-Bench Creative Writing | 1611 | — |
| WildBench | — | 80.7% |
| EQ-Bench 4 | 1226 | — |
Frequently asked questions
Is Inkling better than Qwen3-Next 80B-A3B Instruct?
Inkling is the stronger model overall, scoring 44.1 to 43.0 on the Noometry Index. Qwen3-Next 80B-A3B Instruct costs 2.9× 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-Next 80B-A3B Instruct?
Qwen3-Next 80B-A3B Instruct is cheaper. It lists at $0.50 per million input tokens and $2 per million output tokens; Inkling lists at $1.87 and $4.68.
Is Inkling or Qwen3-Next 80B-A3B Instruct better for coding?
Qwen3-Next 80B-A3B Instruct scores higher on coding benchmarks: 42.5 versus 34.5 in the Noometry coding category.
Which has the bigger context window?
Qwen3-Next 80B-A3B Instruct does, with 131K tokens against 66K.
How many benchmarks do Inkling and Qwen3-Next 80B-A3B Instruct share?
17 benchmarks have published results for both models. Inkling has 41 scored results on Noometry and Qwen3-Next 80B-A3B Instruct has 25.