AI Desk · Model IdentityNew arrival
| Signal | Result | Reference |
|---|---|---|
| Tokenizer signature | Not measured | – |
| API surface | No other entry shares this exact contract | api_3b8e24fc |
| Context and output | 1,048,576 context · 384,000 max output | declared |
| Reasoning contract | Optional · efforts max, high, low · default high | declared |
| Serving providers observed | None recorded yet | – |
DeepSeek V4.1 Flash is a sparse mixture-of-experts model from DeepSeek, and the first built on the company's Causal Encoder-Decoder (CED) architecture. It activates 8B parameters on input and 16B on...
Supported parameters
frequency_penaltyinclude_reasoninglogit_biaslogprobsmax_tokensmin_ppresence_penaltyreasoningreasoning_effortrepetition_penaltyresponse_formatseedstopstructured_outputstemperaturetool_choicetoolstop_ktop_logprobstop_p
defaults: {}
Cite this page as the evidence record for DeepSeek: DeepSeek V4.1 Flash: the URL is stable, measurements are dated, and revisions append to the history above. Method: tokenizer fingerprinting, v1, 50 strings.
