---
license: other
tags:
- dflash
- speculative-decoding
- laguna
base_model:
- poolside/Laguna-S-2.1
---

# Laguna-S-2.1-DFlash-polishing

DFlash speculator (drafter) for Laguna-S-2.1, bf16.

- Architecture: `DFlashLagunaForCausalLM` (6 sliding-attention layers, block_size 16).
- Shares token embedding + lm_head with the target; `draft_vocab_size == vocab_size` (no d2t/t2d).
- Source checkpoint: `s3://poolside.titan.checkpoints.us-east-2/adam/checkpoints/ft_sft_e0630_rhiemann_baseline_titan_sft_training/0008400/DFlash_Stage2/25000`
- Loads under vLLM (native `laguna_dflash`) and TRT-LLM (pytorch DFlash backend) as the draft model in a speculative config.