GLM-5.2-NVFP4-dspark

A DSpark speculative-decoding head for nvidia/GLM-5.2-NVFP4, trained with the Speculators library.

DSpark is a semi-autoregressive drafter: it proposes a whole block of tokens in a single forward pass, with a lightweight sequential head injecting intra-block dependencies.

Training

Our training recipe differs from the original DSpark recipe. We use a number of modifications, among them a per-layer fusion mixture over the tapped target layers and an additional deep-layer tap, together with an extended on-policy training schedule. The resulting head uses a 16-token draft block.

Usage

vllm serve nvidia/GLM-5.2-NVFP4 \
  --speculative-config '{"method":"dspark","model":"makora-ai/GLM-5.2-NVFP4-dspark","num_speculative_tokens":15}' \
  --tensor-parallel-size 4 \
  --trust-remote-code

num_speculative_tokens should be 15 (block size 16 = anchor + 15 draft slots).

Validation

Accepted length on the held-out validation split, by training epoch:

epoch	accepted length
1	3.697
2	4.321
3	4.600
4	4.743
5	4.788
6	4.825
7	4.903
8	4.911
9	4.915

Limitations

• Trained and evaluated against nvidia/GLM-5.2-NVFP4. Other quantisations of GLM-5.2 are untested and may accept at a lower rate.
• Validated on B200; other hardware is unvalidated.
• The checkpoint ships a confidence head. Confidence-scheduled verification is not currently implemented in vLLM, so those weights are unused at inference time.