DeepSeek V4.1 Flash

DeepSeek V4.1 Flash is a model for reasoning and agentic coding, part of the deepseek-flash line. It is the first version of its line in the registry. Context window is 1,000,000 tokens, max output 393,216 tokens. Open weights. Supports reasoning mode, tool calling, accepts image and text, returns text. Price from $0.15 per million input tokens and $0.60 per million output tokens, cached input from $0.003 per million tokens.

First version of its line in the registry.

• ·1M token context window
• ·393K max output tokens
• ·Open weights
• ·Reasoning and tool calling
• ·Accepts image and text input

• Reach for it for long-context reasoning tasks
• Reach for it for agentic coding workflows
• Reach for it for multimodal input with text output

3 hosts serve this model at the price above · the maker's documentation

• 8The Goblin Warren: an AI Dungeon Master that never invents a rule2026-10-04

1 signal of the archive mention this model, each scored against the published bar. This list is rebuilt from the archive at every build, so it cannot go stale.

The registry keeps the six most recent versions of a line. Older ones stay in the file and can be surfaced without a rebuild.

What is the context window?

1,000,000 tokens.

What is the price?

From $0.15 per million input tokens and $0.60 per million output tokens, cached input from $0.003 per million.

Are the weights open?

Yes.