Claude Sonnet 5 (Low) vs gpt-oss-20b (High)
Which is better for the money: Claude Sonnet 5 (Low) by Anthropic or gpt-oss-20b (High) by OpenAI? Intelligence, total benchmark cost, and end-to-end time from Artificial Analysis & other sources, snapshot 2026-10-25.
Trade-off
Verdict
Depends on your budget
Claude Sonnet 5 (Low) scores 15.3 points higher (24.3 vs 9.0) but costs 23× as much to run the benchmark ($653 vs $28.8; premium vs budget cost classes). They are not alternatives at one budget: pick Claude Sonnet 5 (Low) when you need intelligence above 9.0, and gpt-oss-20b (High) when 9.0 is enough.
| Metric | Claude Sonnet 5 (Low) | gpt-oss-20b (High) | Difference |
|---|---|---|---|
| Intelligence (higher is better) | 24.3 | 9.0 | −63.0% gpt-oss-20b (High) is 63.0% less intelligent |
| Total AA cost (lower is better) | $652.70 | $28.81 | −95.6% gpt-oss-20b (High) is 95.6% cheaper |
| End-to-end time (lower is better) | 8.46s | 15.23s | +80.0% gpt-oss-20b (High) is 80.0% slower |
| Price per 1M tokens (in / out) | $2.00 / $10.0 | $0.05 / $0.22 | |
| Input modalities | Text, Image | Text |
gpt-oss-20b (High) vs Claude Sonnet 5 (Low): 63.0% less intelligence, 95.6% less cost to run the AA benchmark, and 80.0% slower end-to-end.
Today's Pareto frontier
Against the frontier
- Claude Sonnet 5 (Low) is off the frontier: at the same or lower cost, GPT-6.1 Sol (High) scores 50.20 for $521.
- gpt-oss-20b (High) is off the frontier: at the same or lower cost, GPT-6 Luna (Low) scores 21.50 for $10.6.
Claude Sonnet 5 (Low)gpt-oss-20b (High)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/claude-sonnet-5-low/gpt-oss-20b · Markdown: /api/v1/compare/claude-sonnet-5-low/gpt-oss-20b?format=md · llms.txt · agent skill