Model comparison
Claude Opus 4.8 vs Qwen3.5 27B
Claude Opus 4.8 is the stronger model overall, scoring 60.7 to 41.9 on the Noometry Index. Qwen3.5 27B costs 12× less per token, which makes it the better buy when Claude Opus 4.8'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. Claude Opus 4.8 scores higher in 9 categories and Qwen3.5 27B in 0 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where Claude Opus 4.8 leads 78.4 to 38.8.
- The biggest single-benchmark swing is WeirdML: 82.9% for Claude Opus 4.8 and 39.5% for Qwen3.5 27B.
- Qwen3.5 27B is cheaper at $0.30 / $2.40 per million input/output tokens, against $5 / $25 for Claude Opus 4.8.
- Claude Opus 4.8 accepts more context: 1M tokens versus 262K.
- Qwen3.5 27B has downloadable open weights; the other is API-only.
Side by side
| Claude Opus 4.8 | Qwen3.5 27B | |
|---|---|---|
| Provider | Anthropic | Alibaba (Qwen) |
| Noometry Index | 60.7 | 41.9 |
| Released | 2026-05-28 | 2026-02-23 |
| Weights | Proprietary | Open |
| Context window | 1M | 262K |
| Max output | 128K | 66K |
| Input $ / M tokens | $5 | $0.30 |
| Output $ / M tokens | $25 | $2.40 |
| Results tracked | 65 | 28 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Claude Opus 4.8 leads
Claude Opus 4.8: 59.9 (#12), Qwen3.5 27B: 38.9 (#168)
| Benchmark | Claude Opus 4.8 | Qwen3.5 27B |
|---|---|---|
| LMArena WebDev | 1556 | 1358 |
| WeirdML | 82.9% | 39.5% |
| LMArena Coding | 1490 | 1427 |
| ALE-Bench | 1,564 | 349.45 |
| DeepSWE | 59% | — |
| FrontierCode | 46.5% | — |
| SciCode | 53.5% | — |
| GSO | 47.1% | — |
Agentic & Tool Use Not comparable
Claude Opus 4.8: 47.6 (#11), Qwen3.5 27B: —
| Benchmark | Claude Opus 4.8 | Qwen3.5 27B |
|---|---|---|
| Vending-Bench 2 | 5,787 | 201.98 |
| APEX-Agents | 48.9% | — |
| OSWorld 2.0 | 20.6% | — |
| Remote Labor Index | 8.3% | — |
| τ²-bench Banking | 39.7% | — |
| DeepResearch Bench | 50.2% | — |
| PostTrainBench | 33.8% | — |
| GBAEval | 70.9% | — |
| GDP.pdf | 24% | — |
| LMArena Search | 1204 | — |
Reasoning Claude Opus 4.8 leads
Claude Opus 4.8: 64.7 (#16), Qwen3.5 27B: 27.5 (#117)
| Benchmark | Claude Opus 4.8 | Qwen3.5 27B |
|---|---|---|
| NYT Connections (extended) | 91.1% | 47.9% |
| LMArena Hard Prompts | 1482 | 1414 |
| DTBench | 94.9% | 82.4% |
| LMCA | 57.5% | 34% |
| ARC-AGI-2 | 72.1% | — |
| SimpleBench | 64.8% | — |
| Kagi LLM Benchmark | 88.8% | — |
| ARC-AGI-1 | 92.5% | — |
| CritPt | 20.9% | — |
| Chess Puzzles | 34% | — |
| EnigmaEval | 23.5% | — |
| Thematic Generalization | — | 45.5% |
| EBR-Bench | 28.6% | — |
| Mystery Game Puzzles | 36% | — |
| Surface Evolver Bench | 87.5% | — |
| Bench to the Future 3 | 0.14 | — |
| Epoch Capabilities Index | 158.21 | — |
| ForecastBench | 59.9 | — |
Math Claude Opus 4.8 leads
Claude Opus 4.8: 78.4 (#13), Qwen3.5 27B: 38.8 (#127)
| Benchmark | Claude Opus 4.8 | Qwen3.5 27B |
|---|---|---|
| MathArena Final-Answer Competitions | 91.8% | 56.7% |
| LMArena Math | 1487 | 1429 |
| FrontierMath (Tiers 1-3) | 80% | — |
| FrontierMath Tier 4 | 56.1% | — |
| OTIS Mock AIME 2024-2025 | 98.3% | — |
| ProofBench | 69% | — |
| FrontierMath (Feb 2025 set) | 47.2% | — |
| FrontierMath Tier 4 (v1) | 31.3% | — |
Knowledge Claude Opus 4.8 leads
Claude Opus 4.8: 61.3 (#29), Qwen3.5 27B: 38.0 (#150)
| Benchmark | Claude Opus 4.8 | Qwen3.5 27B |
|---|---|---|
| LMArena Expert | 1502 | 1428 |
| GPQA Diamond | 91% | — |
| SimpleQA Verified | 53% | — |
| Vectara Hallucination Rate | — | 12.1% |
Multimodal Claude Opus 4.8 leads
Claude Opus 4.8: 42.9 (#26), Qwen3.5 27B: 39.4 (#59)
| Benchmark | Claude Opus 4.8 | Qwen3.5 27B |
|---|---|---|
| LMArena Vision | 1294 | 1241 |
| Blueprint-Bench 2 | 14.5% | — |
| Furniture Assembly | 42.5% | — |
| LMArena Document | 1475 | — |
Multilingual Claude Opus 4.8 leads
Claude Opus 4.8: 55.2 (#33), Qwen3.5 27B: 50.8 (#115)
| Benchmark | Claude Opus 4.8 | Qwen3.5 27B |
|---|---|---|
| LMArena Non-English | 1450 | 1390 |
| LMArena Chinese | 1507 | 1478 |
| LMArena French | 1481 | 1410 |
| LMArena German | 1472 | 1393 |
| LMArena Japanese | 1440 | 1345 |
| LMArena Korean | 1432 | 1358 |
| LMArena Russian | 1474 | 1390 |
| LMArena Spanish | 1466 | 1407 |
Instruction Following Claude Opus 4.8 leads
Claude Opus 4.8: 77.4 (#24), Qwen3.5 27B: 73.5 (#119)
| Benchmark | Claude Opus 4.8 | Qwen3.5 27B |
|---|---|---|
| LMArena Instruction Following | 1476 | 1393 |
Long Context Claude Opus 4.8 leads
Claude Opus 4.8: 45.4 (#35), Qwen3.5 27B: 43.1 (#106)
| Benchmark | Claude Opus 4.8 | Qwen3.5 27B |
|---|---|---|
| LMArena Longer Query | 1483 | 1413 |
Writing & Preference Claude Opus 4.8 leads
Claude Opus 4.8: 72.0 (#16), Qwen3.5 27B: 59.3 (#111)
| Benchmark | Claude Opus 4.8 | Qwen3.5 27B |
|---|---|---|
| LMArena Text | 1461 | 1409 |
| LMArena Creative Writing | 1454 | 1362 |
| LMArena Multi-Turn | 1476 | 1410 |
| EQ-Bench Creative Writing | 1840 | — |
| EQ-Bench 4 | 1281 | — |
Frequently asked questions
Is Claude Opus 4.8 better than Qwen3.5 27B?
Claude Opus 4.8 is the stronger model overall, scoring 60.7 to 41.9 on the Noometry Index. Qwen3.5 27B costs 12× less per token, which makes it the better buy when Claude Opus 4.8's lead doesn't matter for your workload.
Which is cheaper, Claude Opus 4.8 or Qwen3.5 27B?
Qwen3.5 27B is cheaper. It lists at $0.30 per million input tokens and $2.40 per million output tokens; Claude Opus 4.8 lists at $5 and $25.
Is Claude Opus 4.8 or Qwen3.5 27B better for coding?
Claude Opus 4.8 scores higher on coding benchmarks: 59.9 versus 38.9 in the Noometry coding category.
Which has the bigger context window?
Claude Opus 4.8 does, with 1M tokens against 262K.
How many benchmarks do Claude Opus 4.8 and Qwen3.5 27B share?
26 benchmarks have published results for both models. Claude Opus 4.8 has 65 scored results on Noometry and Qwen3.5 27B has 28.