Qwen3.8-27BvsClaude Sonnet 5
Across 5 shared benchmarks, Claude Sonnet 5 leads overall: Qwen3.8-27B wins 1, Claude Sonnet 5 wins 4, with 0 ties and an average score difference of -8.81.
Qwen3.8-27B
阿里巴巴 · 2026-08-14 · Reasoning model
Claude Sonnet 5
Anthropic · 2026-06-30 · Multimodal model
Qwen3.8-27B1 win(20%)(80%)4 winsClaude Sonnet 5
Benchmark scores
Grouped by capability, sorted by largest gap within each. 5 shared benchmarks.
AI Agent - Tool Usage
Even 2/2| Benchmark | Qwen3.8-27B | Claude Sonnet 5 | Diff |
|---|---|---|---|
| Terminal-Bench 2.1 | 7327 / 41Thinking (With Tools) | 80.4016 / 41极高强度思考（工具） | -7.40 |
| OSWorld-Verified | 84.303 / 26Thinking (With Tools) | 81.206 / 26极高强度思考（工具） | +3.10 |
Coding and Software Engineer
Claude Sonnet 5 1/1| Benchmark | Qwen3.8-27B | Claude Sonnet 5 | Diff |
|---|---|---|---|
| DeepSWE | 42.2020 / 25Thinking (With Tools) | 5414 / 25Deep Thinking (With Tools) | -11.80 |
General Evaluation
Claude Sonnet 5 1/1| Benchmark | Qwen3.8-27B | Claude Sonnet 5 | Diff |
|---|---|---|---|
| GPQA Diamond | 89.2047 / 226Thinking (No Tools) | 90.5333 / 226极高强度思考（无工具） | -1.33 |
General Knowledge
Claude Sonnet 5 1/1| Benchmark | Qwen3.8-27B | Claude Sonnet 5 | Diff |
|---|---|---|---|
| HLE | 30.8087 / 179Thinking (No Tools) | 57.409 / 179极高强度思考（工具） | -26.60 |
Specs
| Field | Qwen3.8-27B | Claude Sonnet 5 |
|---|---|---|
| Publisher | 阿里巴巴 | Anthropic |
| Release date | 2026-08-14 | 2026-06-30 |
| Model type | Reasoning model | Multimodal model |
| Architecture | Dense | Dense |
| Parameters | 27B | Not available |
| Context length | 256K | 1M |
| Max output | 128K | 128K |
API pricing
Prices use DataLearner records when available; missing fields are not inferred.
| Item | Qwen3.8-27B | Claude Sonnet 5 |
|---|---|---|
| Text input | Not public | $2 / 1M tokens |
| Text output | Not public | $10 / 1M tokens |
| Cache read | Not public | $0.3 / 1M tokens |
| Cache write | Not public | $3.75 / 1M tokens |
One or both models have incomplete public pricing.
Summary
- Claude Sonnet 5leads in:Coding and Software Engineer (1/1), General Evaluation (1/1), General Knowledge (1/1)
- Tied in:AI Agent - Tool Usage
On average across the 5 shared benchmarks, Claude Sonnet 5 scores 8.81 higher.
Largest single-benchmark gap: HLE — Qwen3.8-27B 30.80 vs Claude Sonnet 5 57.40 (-26.60).
Page generated from structured model, pricing and benchmark records. No real-time LLM is used to write the prose.