{ "schema_version": 1, "builder_plan": { "scenario": "A self-hosted assistant is too expensive because its model uses most of the GPU memory.", "change": "Try the paper's supported lower-precision weights on a copy of your serving setup.", "baseline": "The same model at its original precision, on the same GPU.", "steps": [ "Locate the paper's implementation and check its license, model support, data needs, and hardware requirements. Stop here if you cannot obtain the required artifacts.", "Save a small set of representative product tasks and expected outcomes. Keep evaluation tasks out of any training or tuning, and choose quality and cost limits before running.", "Run the baseline and the proposed change on identical inputs. Keep other settings fixed and repeat paired runs to expose variation.", "Save per-task outputs, failures, timings, and all usage costs. Review regressions before expanding the test." ], "measure": "Peak memory, task pass rate, latency, and total serving cost at the same context and load.", "decision": "Keep the change only if it meets your preselected quality and latency limits and improves the metric you care about after setup and operating costs. Otherwise retain the baseline. This is an exploration plan, not a validated recommendation." }, "builder_review": null, "arxiv_id": "2609.38169", "title": "STEPQuant: When and Where Errors Matter in Delta-Rule Recurrent State Quantization", "summary": "STEPQuant replaces full-precision (FP32/INT8) recurrent states in Delta-rule linear-attention models with a post-training, spatial-temporal quantization scheme that allocates bits by error magnitude and memory lifetime, with key-row and value-column scales fitted jointly. It costs only offline calibration plus custom SGLang GPU kernels, adds no inference latency penalty per the reported integration, and yields over 5x state compression with up to 68.7% total serving memory reduction. It can fail if the reader's models use conventional softmax attention with KV caches (where it does not apply), if production state distributions drift from calibration data causing accuracy degradation at low bit-widths, or if the custom kernels do not support the reader's serving stack.", "rationale": "The technique targets only Delta-rule linear-attention architectures such as Qwen3.8-27B and Kimi-Linear-48B, so adoption is warranted only if the reader already serves such models, but the kernel availability and post-training nature make it worth tracking otherwise.", "technique": "Spatial-temporal quantization of Delta-rule recurrent states", "family": "quantizacao", "recommendation": "observar", "published": "2026-09-29", "as_of": "2026-10-18", "editorial_status": "indexed", "source_url": "https://arxiv.org/abs/2609.38169", "independent_implementations": 2, "total_implementations": 3, "stars": 122, "citations": null, "claims": [ { "claim_id": "claim-01", "statement": "STEPQuant replaces full-precision (FP32/INT8) recurrent states in Delta-rule linear-attention models with a post-training, spatial-temporal quantization scheme that allocates bits by error magnitude and memory lifetime, with key-row and value-column scales fitted jointly. It costs only offline calibration plus custom SGLang GPU kernels, adds no inference latency penalty per the reported integration, and yields over 5x state compression with up to 68.7% total serving memory reduction. It can fail if the reader's models use conventional softmax attention with KV caches (where it does not apply), if production state distributions drift from calibration data causing accuracy degradation at low bit-widths, or if the custom kernels do not support the reader's serving stack.", "basis": "inferred", "result": "", "baseline": "", "conditions": "", "source_url": "https://arxiv.org/abs/2609.38169", "source_page": null, "source_excerpt": "" }, { "claim_id": "claim-02", "statement": "6-bit STEPQuant achieves over 5x recurrent-state compression and reduces total serving memory by up to 68.7%, while matching FP32-state accuracy at a nominal 6-bit budget and outperforming uniform INT8 in its 4-bit configuration.", "basis": "inferred", "result": "", "baseline": "", "conditions": "", "source_url": "https://arxiv.org/abs/2609.38169", "source_page": null, "source_excerpt": "" } ], "exposure_map": [ { "dimension": "quality", "basis": "not_evaluated", "finding": "", "interpretation": "", "limitation": "", "next_check": "", "source_url": "", "source_page": null, "source_excerpt": "" }, { "dimension": "compute", "basis": "not_evaluated", "finding": "", "interpretation": "", "limitation": "", "next_check": "", "source_url": "", "source_page": null, "source_excerpt": "" }, { "dimension": "latency", "basis": "not_evaluated", "finding": "", "interpretation": "", "limitation": "", "next_check": "", "source_url": "", "source_page": null, "source_excerpt": "" }, { "dimension": "operations", "basis": "not_evaluated", "finding": "", "interpretation": "", "limitation": "", "next_check": "", "source_url": "", "source_page": null, "source_excerpt": "" }, { "dimension": "compatibility", "basis": "not_evaluated", "finding": "", "interpretation": "", "limitation": "", "next_check": "", "source_url": "", "source_page": null, "source_excerpt": "" }, { "dimension": "security", "basis": "not_evaluated", "finding": "", "interpretation": "", "limitation": "", "next_check": "", "source_url": "", "source_page": null, "source_excerpt": "" }, { "dimension": "data_and_training", "basis": "not_evaluated", "finding": "", "interpretation": "", "limitation": "", "next_check": "", "source_url": "", "source_page": null, "source_excerpt": "" }, { "dimension": "reproducibility", "basis": "not_evaluated", "finding": "", "interpretation": "", "limitation": "", "next_check": "", "source_url": "", "source_page": null, "source_excerpt": "" } ], "risks": [], "minimum_test": [], "open_questions": [], "independent_tests": [], "lab_reproductions": [], "report_available": false, "validation_tier": "unknown", "evidence_tier": "unknown", "infrastructure_basis": "unknown", "equations": [], "equations_status": "not_fetched", "core_kind": "", "equations_fetched_at": "", "equations_source_url": "", "equations_source_sha256": "", "exposures_reviewed_at": "", "exposures_source_sha256": "" }