[2608.11693] Spec Sheets Are Not Kernels: An ISA- and Source-Level Audit of INT8 Availability on NVIDIA Blackwell Ultra
Skip to main content
Search arXiv
Press Enter to search · Advanced search
-->
Computer Science > Hardware Architecture
arXiv:2608.11693 (cs)
[Submitted on 12 Aug 2026]
Title:Spec Sheets Are Not Kernels: An ISA- and Source-Level Audit of INT8 Availability on NVIDIA Blackwell Ultra
Authors:Teng-Ruei Chen<br>View a PDF of the paper titled Spec Sheets Are Not Kernels: An ISA- and Source-Level Audit of INT8 Availability on NVIDIA Blackwell Ultra, by Teng-Ruei Chen
View PDF<br>HTML (experimental)
Abstract:NVIDIA's published specifications give the Blackwell Ultra GPU (B300) a dense-compute ratio of roughly 30:1 between FP8 and INT8 tensor-core throughput; its predecessors, H200 and B200, both provide 1:1. We audit what this deprioritization means in practice by tracing INT8 W8A8 support through four layers of the stack: the published specifications, the PTX ISA, NVIDIA's CUTLASS kernel library, and the two major open-source LLM serving engines (vLLM and SGLang). We find a consistent, layered withdrawal: (i) the PTX ISA never exposes the fifth-generation tensor-core integer path (this http URL with .kind::i8) on sm_103a, even though the same PTX revision extends the FP4 kinds to that target, leaving legacy warp-level IMMA as the only architecturally legal integer tensor-core path on B300; (ii) CUTLASS's kernel generator explicitly skips INT8 UMMA generation for any build targeting 103a, while generating FP8 unconditionally; (iii) vLLM ships no INT8 GEMM for Blackwell and fails with a hard runtime error at the first forward pass, after the model has loaded; and (iv) SGLang's ahead-of-time INT8 GEMM stops at Sm90, while its FP8 tuning configurations already cover B200. We document an escape hatch (rerouting vLLM's INT8 path to a JIT-compiled Triton backend via an environment variable), a false-negative trap in the obvious profiler methodology for detecting "native INT8" on sm_103, and the practical failure semantics that make naive testing expensive. Together, these findings show that a quantization format's availability is a property of the whole stack rather than of the model or the spec sheet. Four distinct layers, three of them NVIDIA's own, withdrew INT8 support in mutually consistent ways, and a format that is nominally present on the datasheet is, by default, undeployable on this hardware.
Comments:<br>8 pages (IEEEtran two-column), 2 figures, 3 tables. All sources pinned to commits/digests with access dates; no performance measurements (companion measurement study in preparation)
Subjects:
Hardware Architecture (cs.AR)
Cite as:<br>arXiv:2608.11693 [cs.AR]
(or<br>arXiv:2608.11693v1 [cs.AR] for this version)
https://doi.org/10.48550/arXiv.2608.11693
Focus to learn more
arXiv-issued DOI via DataCite (pending registration)
Submission history<br>From: Teng-Ruei Chen [view email]<br>[v1]<br>Wed, 12 Aug 2026 06:04:30 UTC (20 KB)
Full-text links:<br>Access Paper:
View a PDF of the paper titled Spec Sheets Are Not Kernels: An ISA- and Source-Level Audit of INT8 Availability on NVIDIA Blackwell Ultra, by Teng-Ruei Chen<br>View PDF<br>HTML (experimental)<br>TeX Source
view license
Current browse context:
cs.AR
next >
new<br>recent<br>| 2026-08
Change to browse by:
cs
References & Citations
NASA ADS<br>Google Scholar
Semantic Scholar
export BibTeX citation<br>Loading...
BibTeX formatted citation
×
loading...
Data provided by:
Bookmark
Bibliographic Tools
Bibliographic and Citation Tools
Bibliographic Explorer Toggle
Bibliographic Explorer (What is the Explorer?)
Connected Papers Toggle
Connected Papers (What is Connected Papers?)
Litmaps Toggle
Litmaps (What is Litmaps?)
scite.ai Toggle
scite Smart Citations (What are Smart Citations?)
Code, Data, Media
Code, Data and Media Associated with this Article
alphaXiv Toggle
alphaXiv (What is alphaXiv?)
Links to Code Toggle
CatalyzeX Code Finder for Papers (What is CatalyzeX?)
DagsHub Toggle
DagsHub (What is DagsHub?)
GotitPub Toggle
Gotit.pub (What is GotitPub?)
Huggingface Toggle
Hugging Face (What is Huggingface?)
ScienceCast Toggle
ScienceCast (What is ScienceCast?)
Demos
Demos
Replicate Toggle
Replicate (What is Replicate?)
Spaces Toggle
Hugging Face Spaces (What is Spaces?)
Spaces Toggle
TXYZ.AI (What is TXYZ.AI?)
Related Papers
Recommenders and Search Tools
Link to Influence Flower
Influence Flower (What are Influence Flowers?)
Core recommender toggle
CORE Recommender (What is CORE?)
Author
Venue
Institution
Topic
About arXivLabs
arXivLabs: experimental projects with community collaborators
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only...