The model DeepSeek-V3.2-Exp-Think is officially named deepseek-reasoner. It is an experimental version. As an intermediate step towards the next-generation architecture, V3.2-Exp introduces DeepSeek Sparse Attention (a sparse attention mechanism) based on V3.1-Terminus, exploring and validating exploratory optimizations for training and inference efficiency on long texts.
Pricing
- Input Tokens: $0.274 /M tokens
- Output Tokens: $0.411 /M tokens
- Cache Read: $0.0274 /M tokens
Input Modalities
- Text
Output Modalities
- Text
Capabilities
- Thinking
- Tools
- Tool calling
- Structured outputs
Providers
DeepSeek deepseek-deepseek-v3.2-exp-think
Pricing$0.274$0.411
Cache$0.0274
Context131K
Max output64K
Latency2.0S
Throughput22.8TPS
Uptime
0.00% uptime 2 days ago
0.00% uptime yesterday
0.00% uptime today
Baidu baidu-deepseek-v3.2-exp-think
Pricing$0.274$0.411
Cache$0.0274
Context128K
Max output16K
Latency2.0S
Throughput22.8TPS
Uptime
0.00% uptime 2 days ago
0.00% uptime yesterday
0.00% uptime today
Performance for DeepSeek-V3.2-Exp-Think
Uptime is the percentage of requests that succeeded over the past 72 hours. AIHubMix continuously monitors every provider and automatically retries with the next-best provider when one returns an error or responds too slowly; Latency is total round-trip time (lower is better); Throughput is how fast the model writes (tokens per second, higher is better).
Uptime
Loading...
Latency
Loading...
Throughput
Loading...
Try this model
Python