All Models
DeepSeek V3.2
DeepSeek-V3.2 is a large language model designed to harmonize high computational efficiency with strong reasoning and agentic tool-use performance. It introduces DeepSeek Sparse Attention (DSA), a fine-grained sparse attention mechanism...
Benchmarks
Available Providers (30)
| Provider | Model ID | Input Cost | Output Cost | Context | Max Output | Docs |
|---|---|---|---|---|---|---|
 Alibaba Token Plan (China) | deepseek-v3.2 | $0/MTok | $0/MTok | 131.1K | 65.5K | |
 Alibaba Token Plan | deepseek-v3.2 | $0/MTok | $0/MTok | 131.1K | 65.5K | |
 CrofAI | deepseek-v3.2 | $0.18/MTok | $0.35/MTok | 163.8K | 163.8K | |
 OpenRouter | deepseek/deepseek-v3.2 | $0.23/MTok | $0.34/MTok | 128K | 64K | |
 Deep Infra | deepseek-ai/DeepSeek-V3.2 | $0.26/MTok | $0.38/MTok | 163.8K | 64K | |
 Kilo Gateway | deepseek/deepseek-v3.2 | $0.26/MTok | $0.38/MTok | 163.8K | 65.5K | |
 Meganova | deepseek-ai/DeepSeek-V3.2 | $0.26/MTok | $0.38/MTok | 164K | 164K | |
 Cortecs | deepseek-v3.2 | $0.27/MTok | $0.44/MTok | 163.8K | 163.8K | |
 NovitaAI | deepseek/deepseek-v3.2 | $0.27/MTok | $0.40/MTok | 163.8K | 65.5K | |
 Helicone | deepseek-v3.2 | $0.27/MTok | $0.41/MTok | 163.8K | 65.5K | |
 Abacus | deepseek-ai/DeepSeek-V3.2 | $0.27/MTok | $0.40/MTok | 128K | 8.2K | |
 Poe | novita/deepseek-v3.2 | $0.27/MTok | $0.40/MTok | 128K | — | |
 Synthetic | hf:deepseek-ai/DeepSeek-V3.2 | $0.27/MTok | $0.40/MTok | 162.8K | 8K | |
 NanoGPT | deepseek/deepseek-v3.2 | $0.28/MTok | $0.42/MTok | 163K | 65.5K | |
 Vercel AI Gateway | deepseek/deepseek-v3.2 | $0.28/MTok | $0.42/MTok | 128K | 8K | |
 Hugging Face | deepseek-ai/DeepSeek-V3.2 | $0.28/MTok | $0.40/MTok | 163.8K | 65.5K | |
 Vivgrid | deepseek-v3.2 | $0.28/MTok | $0.42/MTok | 128K | 128K | |
 ZenMux | deepseek/deepseek-v3.2 | $0.28/MTok | $0.43/MTok | 128K | 64K | |
 LLM Gateway | deepseek-v3.2 | $0.28/MTok | $0.42/MTok | 163.8K | 16.4K | |
 302.AI | deepseek-v3.2 | $0.29/MTok | $0.43/MTok | 128K | 8.2K | |
 Nebius Token Factory | deepseek-ai/DeepSeek-V3.2 | $0.30/MTok | $0.45/MTok | 163K | 16.4K | |
 Venice AI | deepseek-v3.2 | $0.33/MTok | $0.48/MTok | 160K | 32.8K | |
 routing.run | route/deepseek-v3.2 | $0.49/MTok | $0.74/MTok | 163.8K | 163.8K | |
 DigitalOcean | deepseek-3.2 | $0.50/MTok | $1.60/MTok | 128K | 64K | |
 Vultr | nvidia/DeepSeek-V3.2-NVFP4 | $0.55/MTok | $1.65/MTok | 131.1K | 131.1K | |
 Vertex | deepseek-ai/deepseek-v3.2-maas | $0.56/MTok | $1.68/MTok | 163.8K | 65.5K | |
 Azure | deepseek-v3.2 | $0.58/MTok | $1.68/MTok | 128K | 128K | |
 Azure Cognitive Services | deepseek-v3.2 | $0.58/MTok | $1.68/MTok | 128K | 128K | |
 Amazon Bedrock | deepseek.v3.2 | $0.62/MTok | $1.85/MTok | 163.8K | 81.9K | |
 Ollama Cloud | deepseek-v3.2 | —/MTok | —/MTok | 163.8K | 65.5K |
Capabilities
Reasoning
Tool Calling
Attachments
Open Weights
Structured Output