Qwen3.8 27B (Xhigh) vs Qwen3.8 Max (0902)
Which is better for the money: Qwen3.8 27B (Xhigh) by Alibaba or Qwen3.8 Max (0902) by Alibaba? Intelligence, total benchmark cost, and end-to-end time from Artificial Analysis & other sources, snapshot 2026-10-26.
Trade-off
Verdict
Depends on your budget
Qwen3.8 Max (0902) scores 11.7 points higher (45.4 vs 33.7) but costs 3.7× as much to run the benchmark ($4935 vs $1336; flagship vs premium cost classes). They are not alternatives at one budget: pick Qwen3.8 Max (0902) when you need intelligence above 33.7, and Qwen3.8 27B (Xhigh) when 33.7 is enough.
| Metric | Qwen3.8 27B (Xhigh) | Qwen3.8 Max (0902) | Difference |
|---|---|---|---|
| Intelligence (higher is better) | 33.7 | 45.4 | +34.7% Qwen3.8 Max (0902) is 34.7% more intelligent |
| Total AA cost (lower is better) | $1,336.09 | $4,934.79 | +269.3% Qwen3.8 Max (0902) is 3.7× more expensive |
| End-to-end time (lower is better) | 60.72s | 64.07s | +5.5% Qwen3.8 Max (0902) is 5.5% slower |
| Price per 1M tokens (in / out) | $0.50 / $3.00 | $2.00 / $6.00 | |
| Input modalities | Text, Image, Video | Text, Image |
Qwen3.8 Max (0902) vs Qwen3.8 27B (Xhigh): 34.7% more intelligence, 269.3% more cost to run the AA benchmark, and 5.5% slower end-to-end.
Today's Pareto frontier
Against the frontier
- Qwen3.8 27B (Xhigh) is off the frontier: at the same or lower cost, GPT-6.1 Sol (Max) scores 51.80 for $1082.
- Qwen3.8 Max (0902) is off the frontier: at the same or lower cost, Claude Opus 5.5 (Xhigh, Default Fallback) scores 56.00 for $4057.
Qwen3.8 27B (Xhigh)Qwen3.8 Max (0902)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/qwen3-8-27b/qwen3-8-max · Markdown: /api/v1/compare/qwen3-8-27b/qwen3-8-max?format=md · llms.txt · agent skill