DeepSeek: R1 0528 is DeepSeek's model, and its measured tokenizer sits in DeepSeek's own signature group alongside its sibling models. Measured models from StepFun carry the same vocabulary, which reads as reuse of DeepSeek's tokenizer line and questions nobody's authorship; they are listed under "Shares this fingerprint" below.
| Signal | Result | Reference |
|---|---|---|
| Tokenizer signature | Exact match with 12 other models | tk_73c4f0dc |
| API surface | No other entry shares this exact contract | api_7ea3cb80 |
| Context and output | 163,840 context · 32,768 max output | declared |
| Reasoning contract | Mandatory | declared |
| Serving providers observed | StreamLake | measured 3 September 2026 |
Click a row to open the full comparison.
An x marks a row excluded as corrupt (caching or provider interference detected during measurement). Overhead subtracted: 5 prompt tokens.
May 28th update to the [original DeepSeek R1](/deepseek/deepseek-r1) Performance on par with [OpenAI o1](/openai/o1), but open-sourced and with fully open reasoning tokens. It's 671B parameters in size, with 37B active...
Supported parameters
frequency_penaltyinclude_reasoninglogit_biaslogprobsmax_tokensmin_ppresence_penaltyreasoningrepetition_penaltyresponse_formatseedstopstructured_outputstemperaturetool_choicetoolstop_ktop_logprobstop_p
defaults: {"temperature":null,"top_p":null,"frequency_penalty":null}
Cite this page as the evidence record for DeepSeek: R1 0528: the URL is stable, measurements are dated, and revisions append to the history above. Method: tokenizer fingerprinting, v1, 50 strings.
