DeepSeek V3.2 Speciale
DeepSeek
DeepSeek-V3.2-Speciale is a high-compute variant of DeepSeek-V3.2 optimized for maximum reasoning and agentic performance. It builds on DeepSeek Sparse Attention (DSA) for efficient long-context processing, then scales post-training reinforcement learning...
What is DeepSeek V3.2 Speciale?
DeepSeek-V3.2-Speciale is a high-compute variant of DeepSeek-V3.2 optimized for maximum reasoning and agentic performance. It builds on DeepSeek Sparse Attention (DSA) for efficient long-context processing, then scales post-training reinforcement learning...
Specifications
- Developer: DeepSeek
- Context window: 163.8K tokens
- Max output: 163.8K tokens
- Input modalities: text
- Output modalities: text
- Input price: $0.2870 per 1M tokens
- Output price: $0.4310 per 1M tokens
- Knowledge cutoff: —
- Supported parameters: frequency_penalty, include_reasoning, logit_bias, max_tokens, min_p, presence_penalty, reasoning, repetition_penalty, response_format, seed, stop, structured_outputs, temperature, top_k, top_p
Use DeepSeek V3.2 Speciale with PromptLayer
PromptLayer lets teams manage, evaluate, and observe prompts that run on DeepSeek V3.2 Speciale alongside every other model in their stack. Version prompts, run evals across models, and ship safe rollouts from the same dashboard.
Ready to try it yourself? Sign up for PromptLayer and start managing your prompts in minutes.