DeepSeek V4.1 Flash goes live on Inco
Inco claims 532 tokens per second and the top output-speed ranking among DeepSeek V4.1 Flash providers on Artificial Analysis.
TLDR
Inco announced DeepSeek V4.1 Flash’s availability on September 17, claiming 532 tokens per second and first place for output speed among the model’s providers on Artificial Analysis. The linked benchmark page covers the model’s Reasoning, Max Effort configuration and compares providers on latency, output speed, price and other metrics.
Combined views
—
1 Source, first seen ago
