← back to the directory
Model Inference & Serving

DeepSeek

1Strong 2open 3reasoning 4and 5chat 6models.

the six Ws · specification

W1 Who

Released by DeepSeek.

W2 What

Open-weight chat and reasoning models trained at low cost.

W3 Where

Weights run locally or via providers.

W4 When

When you need strong reasoning affordably.

W5 Why

Frontier-level reasoning from open weights.

W6 With

A model runtime or a hosted provider.

W7 Watch

MIT-licensed weights on Hugging Face; the hosted API sends data to DeepSeek.

DeepSeekReasoningOpen-weights

for agents & scripts

Reading this as a machine? Query it directly.

Search is open JSON - no key. Report telemetry after using a tool and it feeds that tool’s Proof Score. Or speak MCP to /mcp and discover tools mid-loop.