gpt-oss-120b (High) vs Qwen3.8 27B (Non-reasoning)
Which is better for the money: gpt-oss-120b (High) by OpenAI 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 8.6 points higher (20.2 vs 11.6) but costs 16× as much to run the benchmark ($2022 vs $123; flagship vs mid-range cost classes). They are not alternatives at one budget: pick Qwen3.8 27B (Non-reasoning) when you need intelligence above 11.6, and gpt-oss-120b (High) when 11.6 is enough.
| Metric | gpt-oss-120b (High) | Qwen3.8 27B (Non-reasoning) | Difference |
|---|---|---|---|
| Intelligence (higher is better) | 11.6 | 20.2 | +74.1% Qwen3.8 27B (Non-reas… is 74.1% more intelligent |
| Total AA cost (lower is better) | $122.71 | $2,022.49 | +1548.2% Qwen3.8 27B (Non-reas… is 16× more expensive |
| End-to-end time (lower is better) | 10.75s | 13.84s | +28.7% Qwen3.8 27B (Non-reas… is 28.7% slower |
| Price per 1M tokens (in / out) | $0.15 / $0.60 | $0.50 / $3.00 | |
| Input modalities | Text | Text, Image, Video |
Qwen3.8 27B (Non-reasoning) vs gpt-oss-120b (High): 74.1% more intelligence, 1548.2% more cost to run the AA benchmark, and 28.7% slower end-to-end.
Today's Pareto frontier
Against the frontier
- gpt-oss-120b (High) is off the frontier: at the same or lower cost, GPT-6 Luna (Max) scores 38.10 for $122.
- 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.
gpt-oss-120b (High)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/gpt-oss-120b/qwen3-8-27b-non-reasoning · Markdown: /api/v1/compare/gpt-oss-120b/qwen3-8-27b-non-reasoning?format=md · llms.txt · agent skill