Ling3-Flash-UBTree-B16

Paper · Target model

Ling3-Flash-UBTree-B16 is a UBTree speculative-decoding draft model for inclusionAI/Ling-3.0-flash. It combines a five-layer DFlash backbone with a learned bigram selector to propose multiple candidate continuations for target-model verification.

SGLang integration — stay tuned! Our UBTree adaptation for Ling3 has not yet been released to the open-source SGLang community. Public installation instructions and serving recipes will be added here when it becomes available.

This checkpoint is used alongside the matching target model; it cannot serve as a standalone text-generation model.

How UBTree works

UBTree pipeline

UBTree pairs a unigram proposer with a bigram selector:

A DFlash backbone proposes candidate tokens for future positions in parallel.
The selector scores transitions between adjacent candidates in parallel.
Tree construction retains promising branches within a verification budget.
The target verifies the tree and commits the accepted continuation.

The paper also introduces tree-native training with renormalized KL supervision on high-temperature target traces, encouraging plausible alternative branches. See the paper for the method and experiments.

Model specifications

Item	Value
Target model	inclusionAI/Ling-3.0-flash
Draft architecture	`Ling3UBTreeModel`
Draft backbone	5 full-attention Transformer layers
Parameters stored in this checkpoint	481,983,774 (482.0M)
Weight dtype	BF16
Hidden size	2,560
Attention heads / KV heads	32 / 32
Target auxiliary feature layers (zero-based)	1, 11, 23, 29, 35
Draft block size	16 (anchor-inclusive runtime convention)
Candidate top-k	64 per draft position
Default verification tree budget	64 non-root nodes + 1 anchor
Training	DFlash warm start, then joint backbone + UBTree selector training

The parameter count covers tensors distributed in this repository. The draft shares token embeddings and the language-model head with the target; these shared weights are not included here. B16 denotes the draft block size, not the verification tree's node budget.

Serving with SGLang

UBTree support for Ling3 in SGLang is coming soon. We will share installation instructions and serving examples once the integration is publicly available. Stay tuned!

Benchmark results

All methods use block size 16, TP4, concurrency 1, BF16, and greedy no-thinking decoding. UBTree uses a 32-node tree.

Each cell shows acceptance length / E2E speedup over target-only AR.

Dataset	UBTree	DFlash	DSpark
GSM8K	9.70 / 3.227×	7.44 / 2.927×	8.00 / 2.899×
MATH-500	9.58 / 3.761×	7.07 / 3.284×	7.69 / 3.252×
AIME 2025	8.18 / 3.497×	5.75 / 2.864×	6.39 / 2.883×
HumanEval	10.40 / 3.479×	7.79 / 3.128×	8.03 / 3.014×
MBPP	9.55 / 3.415×	7.01 / 3.011×	6.89 / 2.730×
LiveCodeBench	7.53 / 3.191×	5.13 / 2.543×	5.53 / 2.484×

AL and throughput use three-round medians; the matched AR baseline uses one round.

Citation

If you use UBTree in your work, please cite:

@article{liang2026ubtree,
  title   = {UBTree: Parallel Tree Drafting via Unigram and Bigram Models for Speculative Decoding},
  author  = {Liang, Chumeng and Wang, Linxuan and Peng, Xinyu and Liu, Huabin and Chen, Yuxin and Liu, Ge and Lin, Guang and Song, Qifan and Li, Jianguo},
  journal = {arXiv preprint arXiv:2609.39972},
  year    = {2026},
  url     = {https://arxiv.org/abs/2609.39972}
}

License

The model weights are released under the MIT License. The pipeline figure is from Liang et al., Figure 2, reproduced without modification under CC BY 4.0.