deepseek-ai/deepseek-v3.2 pricing
Chatdeepseek-ai/deepseek-v3.2 · DeepInfra · context —
Price per 1M tokens
In effect since Oct 15, 2026
| Input (uncached) | $0.26 |
|---|---|
| OutputIncludes reasoning tokens | $0.38 |
| Cache read | $0.13 |
- No long-context tier: one rate for every prompt size.
Price history
1 version since Oct 15, 2026
- Input
- Output
Show data as table
| Effective from | Input / 1M | Output / 1M |
|---|---|---|
| Oct 15, 2026 | $0.26 | $0.38 |
| Oct 16, 2026 | $0.26 | $0.38 |
Cost calculator
Calculated with the same function that AI Cost Map uses for your real events.
Includes reasoning tokens.
Read from the prompt cache.
Cost per call
$0.0007
Cost per month
$71.00
Similar models
Same model type, nearest blended price (3 input : 1 output).
| Model | Input | Output | Context | Blended vs this |
|---|---|---|---|---|
| deepseek-ai/deepseek-v3.2 (this model) DeepInfra | $0.26 | $0.38 | — | — |
| qwen/qwen3-coder-next OpenRouter | $0.12 | $0.80 | — | × 1.0 |
| tencent/hy3-preview OpenRouter | $0.18 | $0.60 | — | × 0.98 |
| fw-glm-5.3-flash Azure OpenAI | $0.188 | $0.625 | — | × 1.0 |
| meta-llama/llama-4-scout-17b-16e-instruct Novita AI | $0.18 | $0.59 | — | × 0.97 |
| ~deepseek/deepseek-v4-flash-latest OpenRouter | $0.0139 | $1.152 | — | × 1.0 |
| qwen/qwen2.5-7b-instruct-turbo Together AI | $0.30 | $0.30 | — | × 1.0 |