DeepSeek V3 0324 vs gpt-oss-20b (High)
Which is better for the money: DeepSeek V3 0324 by DeepSeek or gpt-oss-20b (High) by OpenAI? Intelligence, total benchmark cost, and end-to-end time from Artificial Analysis & other sources, snapshot 2026-10-26.
Clear winner
Verdict
Pick DeepSeek V3 0324
gpt-oss-20b (High) vs DeepSeek V3 0324: 7.2% less intelligence, 45.0% more cost to run the AA benchmark, and 240.5% slower end-to-end.
| Metric | DeepSeek V3 0324 | gpt-oss-20b (High) | Difference |
|---|---|---|---|
| Intelligence (higher is better) | 9.7 | 9.0 | −7.2% gpt-oss-20b (High) is 7.2% less intelligent |
| Total AA cost (lower is better) | $19.87 | $28.81 | +45.0% gpt-oss-20b (High) is 45.0% more expensive |
| End-to-end time (lower is better) | 4.39s | 14.95s | +240.5% gpt-oss-20b (High) is 240.5% slower |
| Price per 1M tokens (in / out) | $0.24 / $0.90 | $0.05 / $0.22 | |
| Input modalities | Text | Text |
Today's Pareto frontier
Against the frontier
- DeepSeek V3 0324 is off the frontier: at the same or lower cost, GPT-6 Luna (Low) scores 21.50 for $10.6.
- gpt-oss-20b (High) is off the frontier: at the same or lower cost, GPT-6 Luna (Low) scores 21.50 for $10.6.
DeepSeek V3 0324gpt-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/deepseek-v3-0324/gpt-oss-20b · Markdown: /api/v1/compare/deepseek-v3-0324/gpt-oss-20b?format=md · llms.txt · agent skill