Mistral Small 3.1 vs Qwen3.8 27B (Low)
Which is better for the money: Mistral Small 3.1 by Mistral or Qwen3.8 27B (Low) 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 27B (Low) scores 19.1 points higher (26.2 vs 7.1) but costs 42× as much to run the benchmark ($1165 vs $27.8; premium vs budget cost classes). They are not alternatives at one budget: pick Qwen3.8 27B (Low) when you need intelligence above 7.1, and Mistral Small 3.1 when 7.1 is enough.
| Metric | Mistral Small 3.1 | Qwen3.8 27B (Low) | Difference |
|---|---|---|---|
| Intelligence (higher is better) | 7.1 | 26.2 | +269.0% Qwen3.8 27B (Low) is 3.7× more intelligent |
| Total AA cost (lower is better) | $27.82 | $1,164.93 | +4087.4% Qwen3.8 27B (Low) is 42× more expensive |
| End-to-end time (lower is better) | 10.82s | 58.18s | +437.7% Qwen3.8 27B (Low) is 437.7% slower |
| Price per 1M tokens (in / out) | $0.08 / $0.14 | $0.50 / $3.00 | |
| Input modalities | Text, Image | Text, Image, Video |
Qwen3.8 27B (Low) vs Mistral Small 3.1: 269.0% more intelligence, 4087.4% more cost to run the AA benchmark, and 437.7% slower end-to-end.
Today's Pareto frontier
Against the frontier
- Mistral Small 3.1 is off the frontier: at the same or lower cost, GPT-6 Luna (Low) scores 21.50 for $10.6.
- Qwen3.8 27B (Low) is off the frontier: at the same or lower cost, GPT-6.1 Sol (Max) scores 51.80 for $1082.
Mistral Small 3.1Qwen3.8 27B (Low)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/mistral-small-3-1/qwen3-8-27b-low · Markdown: /api/v1/compare/mistral-small-3-1/qwen3-8-27b-low?format=md · llms.txt · agent skill