AI Model Comparison
Side-by-side comparison of pricing, context window, benchmarks, capabilities, and license. Every figure comes from the developer's own pages, last checked on ; each model's page lists its sources. A figure we could not source shows as a dash.
Model A — DeepSeek
DeepSeek-R1 is an open-weight reasoning model from DeepSeek, released on January 20, 2025 under the MIT licence, with 671B total and 37B activated parameters.
Model B — OpenAI
o3 is a reasoning model from OpenAI, released on April 16, 2025 (snapshot o3-2025-04-16). It takes text and image input and returns text.
| Property | DeepSeek-R1 | OpenAI o3 |
|---|---|---|
| Released | Jan 20, 2025 | Apr 16, 2025 |
| Developer | DeepSeek | OpenAI |
| Type | reasoning | reasoning |
| License | open-weight | proprietary |
| Context window | 128,000 | 200,000 |
| Parameters | 671B total, 37B active (mixture of experts) | — |
| Developer status | Retired | Deprecated |
| Input $/M tokens | — | $2.00 |
| Output $/M tokens | — | $8.00 |
Bold green values indicate the winner on that dimension where comparison is meaningful (lower price wins; bigger context wins; more open license wins; newer date wins). Prices are the ones each developer lists today; a model it no longer prices has none.
| Benchmark | DeepSeek-R1 | OpenAI o3 |
|---|---|---|
| SWE-bench Verified | 49.2%DeepSeek’s figure, resolved. | 69.1%OpenAI’s figure, on 477 of the 500 tasks, as reported alongside GPT-5. |
Each score is its developer's own published figure, run under that developer's own test setup, so the two are not always measured the same way. The conditions are under each figure, and neither is marked as ahead unless both were run under the same stated conditions.
No exclusive capabilities
Programmatic access available at /api/ai/compare?a=deepseek-r1&b=o3. CORS is open. Citation guidance: /about/editorial-process.