DeepSeek V4 Pro
DeepSeek's April 2026 flagship. 1.6T MoE / 49B active params, 1M context, rivals top closed-source models on STEM and coding at a fraction of the price.
DeepSeek develops cutting-edge open-weight reasoning models including DeepSeek V3 and DeepSeek R1, rivaling frontier closed models on math, code, and reasoning benchmarks.
Access every DeepSeek model through Railwail's OpenAI-compatible API.
5 models
DeepSeek's April 2026 flagship. 1.6T MoE / 49B active params, 1M context, rivals top closed-source models on STEM and coding at a fraction of the price.
Efficiency-optimized variant of DeepSeek V4. 284B MoE / 13B active, 1M context, ultra-low pricing for high-throughput workloads.
DeepSeek's current Flash model (API name deepseek-flash, model version DeepSeek-V4.1-Flash): 1M-token context, up to 384K output tokens, JSON output, tool calls and vision input.
DeepSeek-VL 7B chat model. Vision-language model with hybrid vision encoder and strong real-world visual question answering performance.
DeepSeek's unified multimodal model. Decouples vision encoding for both understanding and generation tasks.
Listed for reference. These models have no verified price or are no longer served, so the API does not run them.
Free credits on sign-up. No credit card required. Access DeepSeek and 27+ other providers through a single API.