Claude Sonnet 5.5 (Max, Default Fallback) vs gpt-oss-20b (High)

Which is better for the money: Claude Sonnet 5.5 (Max, Default Fallback) 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

AnthropicClaude Sonnet 5.5 (Max, Default Fallback)Off the frontier

Intelligence: 56.00
Total cost: $7260

OpenAIgpt-oss-20b (High)Off the frontier

Intelligence: 9.00
Total cost: $28.8

Verdict

Depends on your budget

Claude Sonnet 5.5 (Max, Default Fallbac… scores 47.0 points higher (56.0 vs 9.0) but costs 252× as much to run the benchmark ($7260 vs $28.8; flagship vs budget cost classes). They are not alternatives at one budget: pick Claude Sonnet 5.5 (Max, Default Fallbac… when you need intelligence above 9.0, and gpt-oss-20b (High) when 9.0 is enough.

Metric	Claude Sonnet 5.5 (Max, Default Fallback)	gpt-oss-20b (High)	Difference
Intelligence (higher is better)	56.0	9.0	−83.9% gpt-oss-20b (High) is 83.9% less intelligent
Total AA cost (lower is better)	$7,259.50	$28.81	−99.6% gpt-oss-20b (High) is 99.6% cheaper
End-to-end time (lower is better)	380.92s	15.23s	−96.0% gpt-oss-20b (High) is 96.0% faster
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 5.5 (Max, Default Fallback): 83.9% less intelligence, 99.6% less cost to run the AA benchmark, and 96.0% faster end-to-end.

Open this matchup in the interactive picker Claude Sonnet 5.5 (Max, Default Fallback): what beats it gpt-oss-20b (High): what beats it

Today's Pareto frontier

Against the frontier

Claude Sonnet 5.5 (Max, Default Fallback) is off the frontier: at the same or lower cost, Claude Opus 5.5 (Xhigh, Default Fallback) scores 56.00 for $4057.
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 5.5 (Max, Default Fallback)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-5-5/gpt-oss-20b · Markdown: /api/v1/compare/claude-sonnet-5-5/gpt-oss-20b?format=md · llms.txt · agent skill