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 designed …
Model details →DeepSeek: DeepSeek V3.2 Exp vs Inception: Mercury
DeepSeek: DeepSeek V3.2 Exp wins on 3 of 6 axes — pricing and capability skew in its favour for most workloads.
Mercury is the first diffusion large language model (dLLM). Applying a breakthrough discrete diffusion approach, the model runs 5-10x faster than even speed optimized models like GPT-4.1 Nano and Claude 3.5 Haiku while matching their perfor…
Model details →Side-by-side comparison
| Capability | DeepSeek: DeepSeek V3.2 Exp | Inception: Mercury | Winner |
|---|---|---|---|
| Context window Maximum number of input tokens the model can attend to in a single request. | 164K | 128K | 🏆 DeepSeek: DeepSeek V3.2 Exp |
| Input price (per 1M) Cost per million input tokens billed by the provider. | $0.27 | $0.25 | 🏆 Inception: Mercury |
| Output price (per 1M) Cost per million output tokens billed by the provider. | $0.41 | $0.75 | 🏆 DeepSeek: DeepSeek V3.2 Exp |
| Tool / function calling First-class support for emitting structured tool calls. | Yes | Yes | Tie |
| Vision input Accepts image inputs alongside text. | — | — | Tie |
| Reasoning mode Internal chain-of-thought / extended-thinking support. | Yes | — | 🏆 DeepSeek: DeepSeek V3.2 Exp |
Frequently asked questions
Is DeepSeek: DeepSeek V3.2 Exp better than Inception: Mercury?
DeepSeek: DeepSeek V3.2 Exp wins on 3 of 6 axes — pricing and capability skew in its favour for most workloads.
What's the price difference between DeepSeek: DeepSeek V3.2 Exp and Inception: Mercury?
Input: $0.27 vs $0.25 per 1M tokens. Output: $0.41 vs $0.75 per 1M tokens.
What context windows do DeepSeek: DeepSeek V3.2 Exp and Inception: Mercury support?
DeepSeek: DeepSeek V3.2 Exp supports up to 164K tokens. Inception: Mercury supports up to 128K tokens.
Do both DeepSeek: DeepSeek V3.2 Exp and Inception: Mercury support tool calling?
DeepSeek: DeepSeek V3.2 Exp: yes. Inception: Mercury: yes.