Downloads · 30 days
71
20% of all-time downloads
Qwen3-8B-speculator.dspark-reasoning
mgoin/Qwen3-8B-speculator.dspark-reasoning
Qwen3-8B-speculator.dspark-reasoning is a text generation model from mgoin. Use it when you need the model to write or continue text. It is set up for speculators.
DSpark draft model (DFlash backbone + Markov logit-bias head + confidence head) for Qwen/Qwen3-8B, trained with speculators (PR 677).
All-time downloads
349
Public
Parameters
1.8B
3.7 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors3.7 GB · 100%
Parameter types
How the weights are stored.
BF161.8B · 100%
Try a prompt
Base models
Model card
From the Hugging Face model README
Qwen3-8B DSpark speculator (reasoning recipe)
DSpark draft model (DFlash backbone + Markov logit-bias head + confidence head) for
Qwen/Qwen3-8B, trained with speculators
(PR #677).
Trained to match the recipe of the reference DFlash speculator
RedHatAI/Qwen3-8B-speculator.dflash:
ultrachat answers regenerated by Qwen/Qwen3-8B with reasoning, sequence length 8192,
lr 6e-4, cosine schedule, silu, max_anchors 3072, 3 epochs (50k samples).
DSpark adds a Markov logit-bias head (intra-block token dependency) and a per-position confidence head on top of the DFlash backbone; the matched-recipe DFlash above is the baseline reference.
Validation per-position acceptance (pos 1-7): 0.711 / 0.590 / 0.506 / 0.445 / 0.401 / 0.367 / 0.340 | full_acc 0.480 | confidence ECE-style abs error 0.054.
- Task
- Text Generation
- Library
- speculators
- Created
- Jun 28, 2026
- Updated
- Jun 28, 2026