EvSpark: Lossless Speculative Decoding for Hybrid DNA Foundation Models

Hybrid DNA foundation models combine convolutional, recurrent, and attention layers, making speculative decoding more difficult than truncating a KV cache. We present EvSpark , a speculative decoding system for Evo2 that verifies draft blocks in parallel and restores all three classes of inference state by selecting retained intermediate states, without replay. A compact, hidden-state-conditioned drafter proposes each block in one parallel forward pass. On Evo2 7B, a 48-prompt benchmark with three training seeds yields 2.96 × on 43 real-sequence prompts and 3.27 × including five synthetic controls. Acceleration persists at 262k-token context (1.84 × –2.43 × on two bacterial genomes) and over 32k generated tokens. Retraining the same drafter architecture for Evo2 20B and 40B yields 2.18 × –2.46 × on real sequences and 2.51 × –2.78 × on the full suite. Autoregressive drafter comparisons and batch measurements show why low draft latency, rather than acceptance alone, determines the gain. The method preserves the target distribution. Greedy checks find no non-tie divergences across 48 prompts and six checkpoints; complementary sampling diagnostics quantify decoding fidelity. A drafter trained on 30M supervised positions reaches 2.82 × on real sequences, compared with 2.96 × at 150M positions. In regulatory-DNA design, EvSpark achieves a median complete-workflow speedup of 1.57 × over a calibrated batched native baseline.