ReMM SLERP 13B is Undi95's model. Its measured tokenizer sits with the Llama2 group, which is evidence that it builds on the Llama2 tokenizer. Tokenizer reuse is normal engineering, open vocabularies travel between labs, and this says nothing against Undi95's authorship of the model itself.
| Signal | Result | Reference |
|---|---|---|
| Tokenizer signature | Exact match with 1 other model | tk_5348f9f6 |
| API surface | No other entry shares this exact contract | api_9ccdd828 |
| Context and output | 6,144 context · 5,529 max output | declared |
| Reasoning contract | None declared | declared |
| Serving providers observed | Mancer 2 | measured 22 August 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: 36 prompt tokens.
A recreation trial of the original MythoMax-L2-B13 but with updated models. #merge
Supported parameters
frequency_penaltylogit_biaslogprobsmax_tokensmin_ppresence_penaltyrepetition_penaltyresponse_formatseedstopstructured_outputstemperaturetop_atop_ktop_logprobstop_p
defaults: {}
Cite this page as the evidence record for ReMM SLERP 13B: the URL is stable, measurements are dated, and revisions append to the history above. Method: tokenizer fingerprinting, v1, 50 strings.
