DeepSeek V4.1
The DeepSeek V4.1 family from DeepSeek AI introduces multimodal Mixture-of-Experts model built for long-context reasoning and agentic workloads. It combines million-token contexts, efficient sparse architectures, controllable reasoning, and native vision-language capabilities.
2 variants
DeepSeek V4.1 Flash 1017
DeepSeek
Params: 763B
DeepSeek V4.1 Flash 1017 is an updated multimodal Mixture-of-Experts model for agentic coding, reasoning, and visual tasks, with shorter outputs and faster task execution. It has 763B total parameters with 8B active during prefill and 16B during decode, using a 40-layer Causal Encoder-Decoder with 5,120 hidden size, 64 attention heads, and 1 KV head. The architecture pairs Compressed Sparse Attention 2 with 384 routed experts and a shared expert, activating 6 per token. Supporting a 1M-token context, it adds a 32-layer vision encoder and DSpark speculative decoding.
Text GenerationInstruction Following+8
DeepSeek V4.1 Flash
DeepSeek
Params: 763B
DeepSeek V4.1 Flash is a multimodal Mixture-of-Experts model designed for long-context reasoning, agentic workflows, and input-heavy workloads across text and images. It has 763B total parameters with 8B active during prefill and 16B during decode, using a 40-layer Causal Encoder-Decoder architecture with 5,120 hidden size, 64 attention heads, and 1 KV head. The model employs Compressed Sparse Attention 2, 384 routed experts with 6 activated per token, 1 shared expert, and 3 Multi-Token Prediction layers, alongside Engram conditional memory and DSpark speculative decoding. Supporting 1M-token context, it uses a 32-layer vision encoder for multimodal processing.
Text GenerationInstruction Following+8