DeepSeek V3.2 Exp
DeepSeek · Open weights · Released Sep 29, 2025 · deepseek/deepseek-v3.2-exp
164K contextToolsReasoningStructured outputPrompt cachingOpen weights
Available from 3 providersUSD / MTOK · VERIFIED 4 MINUTES AGO
EST. MONTHLY COSTclick a row for detail · click headers to sort
| PROVIDER | INPUT | OUTPUT | CACHE | CONTEXT | QUANT | TTFT | TPS | UPTIME |
|---|---|---|---|---|---|---|---|---|
| $0.27 | $0.41 | — | 164K | fp8 | 2.5s | 34 tok/s | 99.3% | |
| $0.27 | $0.41 | — | 164K | fp8 | 3s | 45 tok/s | 87.9% | |
| ATAtlasCloud | $0.27 | $0.41 | $0.27 | 164K | fp8 | 2.7s | 41 tok/s | 99.6% |
TTFT/TPS: median (P50) over trailing 72h · chat10k profile (~10K in / 200 out) · measured via OpenRouter routing · methodology v1.1 · measured 44 hours ago
About
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...
Specifications
Context window163,840
Max output65,536
Modalities intext
Modalities outtext
LicenseOpen weights
ReleasedSep 29, 2025
Knowledge cutoff2025-07-31
TokenizerDeepSeek
WeightsHugging Face ↗
Related
Embed badge
[](https://modelindex.ai/models/deepseek/deepseek-v3.2-exp)Sources
OpenRouter API4 minutes ago
AGGREGATORHugging Face4 minutes ago
COMMUNITYModelIndex benchmarks44 hours ago
MEASUREDVerified 4 minutes ago · list prices, not negotiated rates.