DeepSeek
- Protocol
- deepseek://
- Default model
- deepseek-flash
Runs DeepSeek models on DeepSeek’s OpenAI-compatible chat API at https://api.deepseek.com.
Options
| Option | On DeepSeek |
|---|---|
vision | On by default for deepseek-flash models; vision=true opts another model in |
body. | Fields OpenAI’s chat request names |
header, max_, temperature, top_, frames, timeout, insecure | As listed |
Example
CLI
Run locally
zshot -t llm_text \
--llm-format-model "deepseek://?model=deepseek-flash" \
--llm-format-prompt "Summarize the lead story in one sentence." \
-f summary.txt https://zshot-cli.com/example_assets/gazette/