Qwen3.5 9B (Reasoning) vs Qwen3.8 27B (Non-reasoning)
Which is better for the money: Qwen3.5 9B (Reasoning) by Alibaba or Qwen3.8 27B (Non-reasoning) by Alibaba? Intelligence, total benchmark cost, and end-to-end time from Artificial Analysis & other sources, snapshot 2026-10-27.
Trade-off
Verdict
Depends on your budget
Qwen3.8 27B (Non-reasoning) scores 9.0 points higher (20.2 vs 11.2) but costs 6.2× as much to run the benchmark ($2022 vs $329; flagship vs mid-range cost classes). They are not alternatives at one budget: pick Qwen3.8 27B (Non-reasoning) when you need intelligence above 11.2, and Qwen3.5 9B (Reasoning) when 11.2 is enough.
| Metric | Qwen3.5 9B (Reasoning) | Qwen3.8 27B (Non-reasoning) | Difference |
|---|---|---|---|
| Intelligence (higher is better) | 11.2 | 20.2 | +80.4% Qwen3.8 27B (Non-reas… is 80.4% more intelligent |
| Total AA cost (lower is better) | $328.86 | $2,022.49 | +515.0% Qwen3.8 27B (Non-reas… is 6.2× more expensive |
| End-to-end time (lower is better) | 43.79s | 13.84s | −68.4% Qwen3.8 27B (Non-reas… is 68.4% faster |
| Price per 1M tokens (in / out) | $0.14 / $0.20 | $0.50 / $3.00 | |
| Input modalities | Text, Image, Video | Text, Image, Video |
Qwen3.8 27B (Non-reasoning) vs Qwen3.5 9B (Reasoning): 80.4% more intelligence, 515.0% more cost to run the AA benchmark, and 68.4% faster end-to-end.
Today's Pareto frontier
Against the frontier
- Qwen3.5 9B (Reasoning) is off the frontier: at the same or lower cost, MiMo-V2.6-Pro scores 46.30 for $207.
- Qwen3.8 27B (Non-reasoning) is off the frontier: at the same or lower cost, GPT-6.1 Sol (Max) scores 51.80 for $1082.
Qwen3.5 9B (Reasoning)Qwen3.8 27B (Non-reasoning)Today's Pareto frontierOpen-weights frontierEvery other model
Matchup
Compare something else
Agents:
Accept: text/markdown or ?format=md returns this page as Markdown. JSON: /api/v1/compare/qwen3-5-9b/qwen3-8-27b-non-reasoning · Markdown: /api/v1/compare/qwen3-5-9b/qwen3-8-27b-non-reasoning?format=md · llms.txt · agent skill