A
DSpark
speculative-decoding drafter
trained for the verifier
thinkingmachines/Inkling-Small-NVFP4
(multimodal MoE, NVFP4). DSpark drafts a whole block in one parallel pass
(DFlash-style backbone) and injects intra-block dependency with a lightweight
sequential Markov head, plus a per-position confidence head.
Model details
Algorithm
DSpark (greedy proposal)
Verifier
thinkingmachines/Inkling-Small-NVFP4
Backbone layers
6
Hidden size
4096
Attention heads / KV heads
32 / 8
Draft vocab size
201024 (full padded vocab)
Block size
16
Markov rank
256 (vanilla)
Confidence head
enabled (with Markov)
Aux hidden-state layers
2, 10, 18, 26, 34, 39
dtype
bfloat16
Acceptance evaluation
Inference acceptance measured with vLLM spec-decode over 9 task subsets
(~300k drafts total).
Mean acceptance length (AL)
= expected number of
tokens accepted per draft step (higher is better; block size 16).
Subset
Acceptance length
pos-0
pos-1
pos-2
pos-3
pos-4
pos-5
pos-6
pos-7
math_reasoning
4.55
0.776
0.611
0.476
0.376
0.297
0.232
0.182
0.141
HumanEval
3.90
0.740
0.533
0.389
0.288
0.215
0.165
0.129
0.101
rag
3.73
0.736
0.537
0.394
0.288
0.212
0.156
0.113
0.083
translation
3.56
0.736
0.535
0.382
0.269
0.188
0.132
0.094
0.067
summarization
3.40
0.731
0.530
0.375
0.265
0.178
0.119
0.077
0.050
tool_call
3.20
0.707
0.484
0.328
0.222
0.148
0.100
0.067
0.045
writing
3.11
0.679
0.447
0.296
0.198
0.136
0.095
0.068
0.050
qa
3.10
0.684
0.464
0.318
0.216
0.146
0.098
0.065
0.043
question
3.09
0.679
0.446
0.294
0.197
0.136
0.093
0.067
0.048
Average
3.40
0.709
0.493
0.344
0.241
0.170
0.121
0.087
0.063
Draft-weighted average across all subsets shown in the last row.
Per-position acceptance rate, draft-weighted mean over all subsets:
Required vLLM patches
(both open at time of writing):
vllm-project/vllm#52184
—
adds Eagle3 hidden-state extraction support for Inkling (the verifier must
expose the aux hidden states this drafter consumes).
vllm-project/vllm#55133
—
treats a full/padded-vocab draft (draft
vocab_size
201024 vs. the verifier's
logical 200058) as an identity map, so no
d2t
mapping is required to load it.
Also note: Inkling currently requires
--enforce-eager
(CUDA-graph capture of
the custom ops crashes).
Training
Trained with
speculators
for 3
epochs (
checkpoint_best
= best validation epoch). Validation at the selected
checkpoint: accept_len 3.67, accept_rate 0.40, pos-0 acc 0.776. Loss
{"ce": 0.1, "tv": 0.9}
, lr 1e-4, seq len 8192, block size 16.
Training curves (run
dspark_inkling_small_v2
, logged with trackio):
Provenance / reproducibility:
train_command.txt
— exact training command, git SHA, and
package versions (speculators 0.7.0.dev155, vllm 0.27.1, transformers 5.15.0,
torch 2.13.0+cu130, compressed-tensors 0.17.0).
Runs of inference-optimization dspark-inkling-small on huggingface.co
87
Total runs
0
24-hour runs
2
3-day runs
7
7-day runs
87
30-day runs
More Information About dspark-inkling-small huggingface.co Model
dspark-inkling-small huggingface.co
dspark-inkling-small huggingface.co is an AI model on huggingface.co that provides dspark-inkling-small's model effect (), which can be used instantly with this inference-optimization dspark-inkling-small model. huggingface.co supports a free trial of the dspark-inkling-small model, and also provides paid use of the dspark-inkling-small. Support call dspark-inkling-small model through api, including Node.js, Python, http.
dspark-inkling-small huggingface.co is an online trial and call api platform, which integrates dspark-inkling-small's modeling effects, including api services, and provides a free online trial of dspark-inkling-small, you can try dspark-inkling-small online for free by clicking the link below.
inference-optimization dspark-inkling-small online free url in huggingface.co:
dspark-inkling-small is an open source model from GitHub that offers a free installation service, and any user can find dspark-inkling-small on GitHub to install. At the same time, huggingface.co provides the effect of dspark-inkling-small install, users can directly use dspark-inkling-small installed effect in huggingface.co for debugging and trial. It also supports api for free installation.
dspark-inkling-small install url in huggingface.co: