Models / DeepSeek / DeepSeek V3.2
DeepSeek · deepseek
DeepSeek V3.2
Hybrid-reasoning DeepSeek model with thinking and non-thinking modes, sparse attention, and tool-use
$0.50 in · $1.00 out
per 1M tokens · provider rate closest to the median. No first-party DeepSeek price was listed. The reference price is NanoGPT's rate, the provider closest to the median.
Provider input prices run from $0.22 to $3.00 per 1M tokens. Output prices run from $0.33 to $4.50.
- Context
- 128,000
- Max output
- 64,000
- Cache read
- $0.25
- Cache write
- —
- Reasoning
- Yes
- Tool calls
- Yes
- Structured output
- Yes
- Vision
- No
- Open weights
- Yes
- License
- MIT License
- Released
- Dec 1, 2025
- Updated in source
- Dec 1, 2025
- Knowledge
- 2024-07
- Providers
- 16
Providers
| Provider | Input | Output | Cache read | Cache write | Status |
|---|---|---|---|---|---|
| TokenGo | $0.22 | $0.33 | $0.06 | — | — |
| Kilo Gateway | $0.26 | $0.80 | $0.14 | — | — |
| OpenRouter | $0.26 | $0.80 | $0.14 | — | — |
| LLM Gateway | $0.27 | $0.40 | $0.13 | — | — |
| Merge Gateway | $0.28 | $0.40 | $0.13 | — | — |
| Ofox | $0.29 | $0.43 | $0.06 | — | — |
| Venice AI | $0.33 | $0.48 | $0.16 | — | — |
| Melious | $0.35 | $0.58 | $0.093 | — | — |
| Cortecs | $0.50 | $1.50 | $0.15 | — | — |
| Friendli | $0.50 | $1.50 | $0.25 | — | — |
| NanoGPTreference | $0.50 | $1.00 | $0.25 | — | — |
| Amazon Bedrock | $0.62 | $1.85 | — | — | — |
| Vercel AI Gateway | $0.62 | $1.85 | — | — | — |
| Chutes | $1.00 | $1.00 | $0.10 | — | — |
| NEAR AI Cloud | $1.10 | $1.10 | $0.55 | — | — |
| SambaNova | $3.00 | $4.50 | — | — | — |
Jobs this model qualifies for
- Coding agents — clears the bar
- Cheapest summarization — clears the bar