DeepSeek V3 (Dec '24) vs gpt-oss-20b (High)
Which is better for the money: DeepSeek V3 (Dec '24) by DeepSeek or gpt-oss-20b (High) by OpenAI? Intelligence, total benchmark cost, and end-to-end time from Artificial Analysis & other sources, snapshot 2026-10-24.
Trade-off
Verdict
Depends on your budget
gpt-oss-20b (High) scores 0.5 points higher (9.0 vs 8.5) but costs 1.1× as much to run the benchmark ($28.8 vs $25.4; both budget class). Pick gpt-oss-20b (High) when the extra intelligence is worth the premium, otherwise DeepSeek V3 (Dec '24).
| Metric | DeepSeek V3 (Dec '24) | gpt-oss-20b (High) | Difference |
|---|---|---|---|
| Intelligence (higher is better) | 8.5 | 9.0 | +5.9% gpt-oss-20b (High) is 5.9% more intelligent |
| Total AA cost (lower is better) | $25.37 | $28.81 | +13.6% gpt-oss-20b (High) is 13.6% more expensive |
| End-to-end time (lower is better) | 11.89s | 16.79s | +41.2% gpt-oss-20b (High) is 41.2% slower |
| Price per 1M tokens (in / out) | $0.32 / $0.89 | $0.05 / $0.22 | |
| Input modalities | Text | Text |
gpt-oss-20b (High) vs DeepSeek V3 (Dec '24): 5.9% more intelligence, 13.6% more cost to run the AA benchmark, and 41.2% slower end-to-end.
Today's Pareto frontier
Against the frontier
- DeepSeek V3 (Dec '24) 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 (Dec '24)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/deepseek-v3/gpt-oss-20b · Markdown: /api/v1/compare/deepseek-v3/gpt-oss-20b?format=md · llms.txt · agent skill