DeepSeek cuts V4.1 Flash prices, raises V4 Pro rates

DeepSeek V4 Pro 0423 now costs $0.9483 per million input tokens and $1.8966 per million output tokens, up from $0.4949 and $0.9897 on 2026-11-08. The input price rose 91.6 percent and the output price rose 91.6 percent. The model has 1598.8B parameters and a context window of 1048576 tokens.

DeepSeek V4.1 Flash 1017 now costs $0.006 per million input tokens and $0.5 per million output tokens, down from $0.15 and $0.6. The input price fell 96.0 percent and the output price fell 16.7 percent. DeepSeek V4.1 Flash (batch) now costs $0.011 per million input tokens and $0.2 per million output tokens, down from $0.112 and $0.336. The input price fell 90.2 percent and the output price fell 40.5 percent. Both flash models have 763.2B parameters and a context window of 1048576 tokens.

deepseek-ai announced DeepSeek-V4-Pro. According to deepseek-ai, DeepSeek-V4-Pro has 1.6T total parameters and 49B activated parameters, and DeepSeek-V4-Flash has 284B total parameters and 13B activated parameters. Both support a context length of one million tokens. According to deepseek-ai, DeepSeek-V4-Pro requires only 27 percent of single-token inference FLOPs compared with DeepSeek-V3.2 in the 1M-token context setting, and requires only 10 percent of KV cache compared with DeepSeek-V3.2 in the 1M-token context setting. According to deepseek-ai, DeepSeek-V4-Pro-Max achieves a Codeforces rating of 3206 and 93.5 Pass@1 on LiveCodeBench.