Model comparison
Claude Sonnet 4 vs Qwen3-30B-A3B
Claude Sonnet 4 is the stronger model overall, scoring 40.6 to 38.5 on the Noometry Index. Qwen3-30B-A3B costs 28× less per token, which makes it the better buy when Claude Sonnet 4's lead doesn't matter for your workload.
Last verified . 29 shared benchmarks.
Alibaba (Qwen)Summary
- They share 29 benchmarks with published results for both. Claude Sonnet 4 scores higher in 6 categories and Qwen3-30B-A3B in 3 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in agentic & tool use, where Claude Sonnet 4 leads 38.3 to 29.5.
- The biggest single-benchmark swing is Kagi LLM Benchmark: 73% for Claude Sonnet 4 and 54.9% for Qwen3-30B-A3B.
- Qwen3-30B-A3B is cheaper at $0.12 / $0.50 per million input/output tokens, against $3 / $15 for Claude Sonnet 4.
- Claude Sonnet 4 accepts more context: 200K tokens versus 41K.
- Qwen3-30B-A3B has downloadable open weights; the other is API-only.
Side by side
| Claude Sonnet 4 | Qwen3-30B-A3B | |
|---|---|---|
| Provider | Anthropic | Alibaba (Qwen) |
| Noometry Index | 40.6 | 38.5 |
| Released | 2025-05-22 | 2025-04-28 |
| Weights | Proprietary | Open |
| Context window | 200K | 41K |
| Max output | 64K | 16K |
| Input $ / M tokens | $3 | $0.12 |
| Output $ / M tokens | $15 | $0.50 |
| Results tracked | 58 | 32 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Claude Sonnet 4 leads
Claude Sonnet 4: 44.0 (#70), Qwen3-30B-A3B: 37.0 (#188)
| Benchmark | Claude Sonnet 4 | Qwen3-30B-A3B |
|---|---|---|
| SciCode | 40% | 33.3% |
| WeirdML | 46.1% | 29.8% |
| LMArena Coding | 1414 | 1416 |
| SWE-bench Verified (bash only) | 64.9% | — |
| Aider Polyglot | 61.3% | — |
| GSO | 4.9% | — |
| ALE-Bench | 655.35 | — |
Agentic & Tool Use Claude Sonnet 4 leads
Claude Sonnet 4: 38.3 (#29), Qwen3-30B-A3B: 29.5 (#75)
| Benchmark | Claude Sonnet 4 | Qwen3-30B-A3B |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 41.4% |
| TheAgentCompany | 33.1% | — |
| Cybench | 35% | — |
| DeepResearch Bench | 46.6% | — |
| OSWorld | 43.9% | — |
| METR Time Horizons | 62% | — |
Reasoning Claude Sonnet 4 leads
Claude Sonnet 4: 23.0 (#169), Qwen3-30B-A3B: 21.6 (#193)
| Benchmark | Claude Sonnet 4 | Qwen3-30B-A3B |
|---|---|---|
| Kagi LLM Benchmark | 73% | 54.9% |
| CritPt | 0.3% | 0.3% |
| LMArena Hard Prompts | 1372 | 1398 |
| DTBench | 77.1% | 69.3% |
| LMCA | 29% | 22.4% |
| Epoch Capabilities Index | 141.69 | 139.63 |
| ARC-AGI-2 | 5.9% | — |
| SimpleBench | 45.5% | — |
| ARC-AGI-1 | 40% | — |
| Chess Puzzles | — | 8% |
| EnigmaEval | 3.1% | — |
| ForecastBench | 60.2 | — |
Math Claude Sonnet 4 leads
Claude Sonnet 4: 41.5 (#71), Qwen3-30B-A3B: 36.0 (#133)
| Benchmark | Claude Sonnet 4 | Qwen3-30B-A3B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 71.1% | 70.3% |
| LMArena Math | 1375 | 1394 |
| MathArena Final-Answer Competitions | — | 47.8% |
| Omni-MATH | 60.2% | — |
| MATH Level 5 | 84.4% | — |
| FrontierMath (Feb 2025 set) | 4.1% | — |
| FrontierMath Tier 4 (v1) | 0% | — |
Knowledge Too close to call
Claude Sonnet 4: 41.6 (#99), Qwen3-30B-A3B: 41.7 (#96)
| Benchmark | Claude Sonnet 4 | Qwen3-30B-A3B |
|---|---|---|
| GPQA Diamond | 79.2% | 70.1% |
| Confabulations | 13.2% | 12.3% |
| LMArena Expert | 1371 | 1395 |
| Humanity's Last Exam | 7.8% | — |
| MMLU-Pro | 84.3% | — |
| Vectara Hallucination Rate | 10.3% | — |
| GPQA (HELM) | 70.6% | — |
Multimodal Not comparable
Claude Sonnet 4: 25.6 (#116), Qwen3-30B-A3B: —
| Benchmark | Claude Sonnet 4 | Qwen3-30B-A3B |
|---|---|---|
| LMArena Vision | 1191 | — |
| GeoBench | 37% | — |
| VPCT | 34% | — |
| MindCube | 44.8% | — |
Multilingual Qwen3-30B-A3B leads
Claude Sonnet 4: 46.9 (#151), Qwen3-30B-A3B: 49.8 (#126)
| Benchmark | Claude Sonnet 4 | Qwen3-30B-A3B |
|---|---|---|
| LMArena Non-English | 1333 | 1372 |
| LMArena Chinese | 1351 | 1434 |
| LMArena French | 1364 | 1419 |
| LMArena German | 1331 | 1379 |
| LMArena Japanese | 1301 | 1338 |
| LMArena Korean | 1293 | 1333 |
| LMArena Russian | 1355 | 1369 |
| LMArena Spanish | 1356 | 1403 |
Instruction Following Too close to call
Claude Sonnet 4: 71.8 (#141), Qwen3-30B-A3B: 72.2 (#137)
| Benchmark | Claude Sonnet 4 | Qwen3-30B-A3B |
|---|---|---|
| LMArena Instruction Following | 1375 | 1364 |
| IFEval | 84% | — |
Long Context Claude Sonnet 4 leads
Claude Sonnet 4: 33.6 (#254), Qwen3-30B-A3B: 31.0 (#278)
| Benchmark | Claude Sonnet 4 | Qwen3-30B-A3B |
|---|---|---|
| Fiction.LiveBench | 46.9% | 40.6% |
| LMArena Longer Query | 1398 | 1379 |
Writing & Preference Claude Sonnet 4 leads
Claude Sonnet 4: 57.3 (#127), Qwen3-30B-A3B: 55.9 (#138)
| Benchmark | Claude Sonnet 4 | Qwen3-30B-A3B |
|---|---|---|
| LMArena Text | 1351 | 1384 |
| LMArena Creative Writing | 1345 | 1318 |
| Short-Story Creative Writing | 81.4% | 75.3% |
| LMArena Multi-Turn | 1376 | 1378 |
| EQ-Bench Creative Writing | 1483 | — |
| WildBench | 83.8% | — |
Frequently asked questions
Is Claude Sonnet 4 better than Qwen3-30B-A3B?
Claude Sonnet 4 is the stronger model overall, scoring 40.6 to 38.5 on the Noometry Index. Qwen3-30B-A3B costs 28× less per token, which makes it the better buy when Claude Sonnet 4's lead doesn't matter for your workload.
Which is cheaper, Claude Sonnet 4 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; Claude Sonnet 4 lists at $3 and $15.
Is Claude Sonnet 4 or Qwen3-30B-A3B better for coding?
Claude Sonnet 4 scores higher on coding benchmarks: 44.0 versus 37.0 in the Noometry coding category.
Which has the bigger context window?
Claude Sonnet 4 does, with 200K tokens against 41K.
How many benchmarks do Claude Sonnet 4 and Qwen3-30B-A3B share?
29 benchmarks have published results for both models. Claude Sonnet 4 has 58 scored results on Noometry and Qwen3-30B-A3B has 32.