DeepSeek
DeepSeek V3.2 Exp
Released 2025-09-29 · Knowledge up to 2025-07-31
DeepSeek-V3.2-Exp is an experimental large language model released by DeepSeek as an intermediate step between V3.1 and future architectures. It introduces DeepSeek Sparse Attention (DSA), a fine-grained sparse attention mechanism...
You send / 1M
$0.270
tokens in
It replies / 1M
$0.410
tokens out
Memory
164K
tokens it can read at once
Typical request
$0.0006
1,000 words in, 400 out
Data privacy
No training or retention: Yes. At least one provider serves it with zero data retention.
| Provider | You send / 1M | It replies / 1M | Uptime | Privacy policy |
|---|---|---|---|---|
| SiliconFlow · fp8 | $0.270 | $0.410 | 100% | Read |
Based on OpenRouter and Vercel AI Gateway data. Confirm terms with the provider before sending sensitive data.
What it can do
- Reads: text
- Writes: text
- Reasoning: Yes
- Tool calling: Yes
- Longest reply: 147K tokens
- Intelligence score: 16.6 / 100 (Artificial Analysis)
Similar models
- DeepSeek V3.2DeepSeek · $0.280 in / $0.420 out per 1M
- GLM 4.7 FlashZ.ai · $0.060 in / $0.400 out per 1M
- Nemotron 3 SuperNVIDIA · $0.080 in / $0.450 out per 1M
- Mercury 3Inception · $0.100 in / $0.350 out per 1M
- LFM2.5-2.6BLiquid · $0.350 in / $0.500 out per 1M
- Qwen3 30B A3BQwen · $0.120 in / $0.500 out per 1M
Appears in
Top No-Training AI ModelsTop No-Training Zero Data Retention AI ModelsTop No-Training Zero Data Retention Cheap AI ModelsTop No-Training Zero Data Retention Reasoning AI ModelsTop No-Training Zero Data Retention Tool-Calling AI ModelsTop No-Training Cheap AI ModelsTop No-Training Cheap Reasoning AI ModelsTop No-Training Cheap Tool-Calling AI Models