Claude 4.5 Sonnet (Reasoning) vs gpt-oss-20b (High)
Which is better for the money: Claude 4.5 Sonnet (Reasoning) 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 4.5 Sonnet (Reasoning) scores 11.7 points higher (20.7 vs 9.0) but costs 52× as much to run the benchmark ($1501 vs $28.8; premium vs budget cost classes). They are not alternatives at one budget: pick Claude 4.5 Sonnet (Reasoning) when you need intelligence above 9.0, and gpt-oss-20b (High) when 9.0 is enough.
| Metric | Claude 4.5 Sonnet (Reasoning) | gpt-oss-20b (High) | Difference |
|---|---|---|---|
| Intelligence (higher is better) | 20.7 | 9.0 | −56.5% gpt-oss-20b (High) is 56.5% less intelligent |
| Total AA cost (lower is better) | $1,501.22 | $28.81 | −98.1% gpt-oss-20b (High) is 98.1% cheaper |
| End-to-end time (lower is better) | 21.58s | 15.23s | −29.4% gpt-oss-20b (High) is 29.4% faster |
| Price per 1M tokens (in / out) | $3.00 / $15.0 | $0.05 / $0.22 | |
| Input modalities | Text, Image | Text |
gpt-oss-20b (High) vs Claude 4.5 Sonnet (Reasoning): 56.5% less intelligence, 98.1% less cost to run the AA benchmark, and 29.4% faster end-to-end.
Today's Pareto frontier
Against the frontier
- Claude 4.5 Sonnet (Reasoning) is off the frontier: at the same or lower cost, GPT-6.1 Sol (Max) scores 51.80 for $1082.
- 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 4.5 Sonnet (Reasoning)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-4-5-sonnet-thinking/gpt-oss-20b · Markdown: /api/v1/compare/claude-4-5-sonnet-thinking/gpt-oss-20b?format=md · llms.txt · agent skill