DeepSeek-V3.2 (Thinking) vs GPT-5.4 mini: Specs & Benchmark Comparison
| Characteristic | DeepSeek-V3.2 (Thinking) | GPT-5.4 mini |
|---|---|---|
| Company | DeepSeek | OpenAI |
| Release Date | November 30, 2025 | March 1, 2026 |
| Parameters | 685B | — |
| Multimodal | No | Yes |
| Context (input) | 131K | 400K |
| Context (output) | 66K | 128K |
| Input Price / 1M | $0.28 | $0.42 |
| Output Price / 1M | $0.42 | $5.50 |
| Average Score | 0.9 | 0.8 |
| Benchmarks | ||
| GPQA | 0.8 | 0.9 |
Visual Benchmark Comparison
DeepSeek-V3.2 (Thinking)
GPT-5.4 mini
GPQA0.8 vs 0.9
0.8
0.9
Verdict
GPT-5.4 mini leads in 2 out of 4 comparison categories.
Overall Performance
Both models show comparable average scores: DeepSeek-V3.2 (Thinking) — 0.9, GPT-5.4 mini — 0.8.
API Cost
DeepSeek-V3.2 (Thinking) is 8.5x cheaper: input $0.28/1M vs $0.42/1M tokens.
Context Window
GPT-5.4 mini supports a larger context: 400K vs 131K tokens.
Recency
GPT-5.4 mini is newer: released 3/1/2026 vs 11/30/2025.
More About These Models
Related Comparisons
The DeepSeek-V3.2 (Thinking) and GPT-5.4 mini comparison is updated for 2026. Data includes benchmark results, API pricing, context window size and other specifications. For more detailed information, visit the DeepSeek-V3.2 (Thinking) or GPT-5.4 mini page. See also the complete list of AI model comparisons.