{
 "schema_version": 1,
 "generated": "2026-10-20",
 "repo": "Dreamer-Toby/STEPQuant",
 "name": "STEPQuant",
 "owner": "Dreamer-Toby",
 "slug": "dreamer-toby-stepquant",
 "github": "https://github.com/Dreamer-Toby/STEPQuant",
 "what": "STEPQuant is Python code that compresses the persistent recurrent states of Delta-rule models by deciding which state units get more bits and how keys and values are scaled, with SGLang serving support.",
 "why": "",
 "language": "Python",
 "requirements": [
  "Model checkpoints for a supported model, such as Qwen3.8-27B, placed on disk",
  "A CUDA toolkit and NVIDIA GPUs (the Qwen example uses four GPUs; the paper used NVIDIA A800 GPUs)",
  "Python with the tested stack: SGLang 0.5.12, PyTorch 2.11.0, Triton 3.7.1",
  "A clone of the repository, run from its root"
 ],
 "audience_note": "Written for ML researchers and inference engineers who run Delta-rule language models on NVIDIA GPUs and want to reproduce or serve the paper's quantized recurrent state.",
 "risk": "none",
 "section": "earlier",
 "weekly_rank": null,
 "category": "Research",
 "url": "https://dedupedai.com/use-cases/dreamer-toby-stepquant/",
 "use_case_count": 3,
 "use_cases": [
  {
   "id": "dreamer-toby-stepquant--reproduce-stepquant6-qwen-results",
   "title": "Reproduce STEP Quant@6 results on Qwen3.8-27B",
   "best": true,
   "url": "https://dedupedai.com/use-cases/dreamer-toby-stepquant/reproduce-stepquant6-qwen-results/",
   "json": "https://dedupedai.com/use-cases/u/dreamer-toby-stepquant/reproduce-stepquant6-qwen-results.json"
  },
  {
   "id": "dreamer-toby-stepquant--serve-quantized-delta-rule-model",
   "title": "Serve a Delta-rule model with compressed recurrent state",
   "best": false,
   "url": "https://dedupedai.com/use-cases/dreamer-toby-stepquant/serve-quantized-delta-rule-model/",
   "json": "https://dedupedai.com/use-cases/u/dreamer-toby-stepquant/serve-quantized-delta-rule-model.json"
  },
  {
   "id": "dreamer-toby-stepquant--compare-quantized-state-accuracy",
   "title": "Compare FP32 and quantized state accuracy on long tasks",
   "best": false,
   "url": "https://dedupedai.com/use-cases/dreamer-toby-stepquant/compare-quantized-state-accuracy/",
   "json": "https://dedupedai.com/use-cases/u/dreamer-toby-stepquant/compare-quantized-state-accuracy.json"
  }
 ],
 "disclaimer": "Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk."
}