vanchin/deepseek-v3
Released by DeepSeek in December 2024, DeepSeek-V3 is a leading open-source Mixture-of-Experts (MoE) language model with 671B total parameters, activating only 37B parameters per token. Pre-trained on 14.8 trillion high-quality tokens, it natively supports a 128K context window. The model leverages an innovative auxiliary-loss-free load balancing strategy, a Multi-head Latent Attention (MLA) architecture, and FP8 mixed-precision training.
Inference Service Provider
The inference service provider for vanchin/deepseek-v3 is Kuaishou Vanchin.
Model Capabilities
| Capability | Support | Capability | Support |
|---|---|---|---|
Input Modality | Text | Output Modality | Text |
Model Experience | Unsupported | Function Calling | Supported |
Structured Outputs | Supported | Web Search | Unsupported |
Prefix Completion | Supported | Context Caching | Supported |
Batch Inference | Unsupported | Fine-tuning | Unsupported |
Context Limits
| Parameter | Value | Parameter | Value |
|---|---|---|---|
Max Input Length | 131072 | Max Output Length | 16384 |
Context Window | 131072 |
Pricing
This page only shows the original pricing for model API calls, excluding any limited-time promotions. Visit Model Studio Console for promotional offers.
China (Beijing)
| Billing Item | Price (CNY) | Unit |
|---|---|---|
Input | 2 | Per 1M tokens |
Output | 8 | Per 1M tokens |
Input(Implicit Cache) | 0.8 | Per 1M tokens |
Rate Limits
China (Beijing)
| Parameter | Value |
|---|---|
RPM (Requests Per Minute) | 500 |
TPM (Tokens Per Minute) | 1,000,000 |