Skip to content
aiqbench takeChips & Infra

aiqbench analysis

This paper addresses a critical infrastructure gap in the AI hardware ecosystem by providing a vendor-neutral catalog of 109 numeric formats and bit-exact conformance vectors. By standardizing reference material for emerging types like FP8, BF16, and MXFP4, it enables engineers to diagnose silent divergences when porting models across different accelerators without relying on proprietary implementations. The work serves as an essential registry tool that clarifies spec-permitted interpretation gaps rather than claiming model accuracy or superiority.
Source coveringarXiv
What this isA model's reading of the article and related coverage, written for aiqbench. Not the publisher's words.

Golden Ruler: A Numeric Format Catalog with Bit-Exact Conformance Vectors for FP8, BF16, MXFP4, and Microscaling Formats

arXiv2 min read

arXiv:2606.09686v3 Announce Type: replace-cross Abstract: Numeric format proliferation in machine learning hardware -- FP8 (E4M3 and E5M2), BF16, MXFP4, microscaling block formats, and dozens of research variants -- has outpaced the availability of vendor-neutral, bit-exact reference material. Engineers porting models across accelerators encounter silent divergences that are difficult to diagnose without a shared ruler. This paper describes a catalog of 109 numeric formats spanning 12 clusters (83 at v2; the count is a catalog invariant, not a fixed number), a suite of six bit-exact conformance packs covering GF16, MXFP4 element, BF16, FP8 E4M3, FP8 E5M2, and E8M0 block scale, and an IEEE P3109 v3.2.0 cross-walk that maps each pack to its corresponding standards-track configured format. Each pack is a self-contained JSON document with a SHA-256 fingerprint, a shared row schema, and an anchor vector that encodes 3.0 -- the identity phi^2 + 1/phi^2 = 3 -- as a cross-pack sanity check. Packs are cross-validated against ml_dtypes 0.5.4 (Google/JAX); any divergence is documented explicitly and interpreted as a spec-permitted interpretation gap rather than hidden. The work is framed as registry filling: it does not propose new formats, make model-accuracy claims, or assert superiority over any vendor's implementation. All artifacts are publicly available at https://github.com/gHashTag/t27 under an open license.

Abstract:Numeric format proliferation in machine learning hardware -- FP8 (E4M3 and E5M2), BF16, MXFP4, microscaling block formats, and dozens of research variants -- has outpaced the availability of vendor-neutral, bit-exact reference material. Engineers porting models across accelerators encounter silent divergences that are difficult to diagnose without a shared ruler. This paper describes a catalog of 109 numeric formats spanning 12 clusters (83 at v2; the count is a catalog invariant, not a fixed number), a suite of six bit-exact conformance packs covering GF16, MXFP4 element, BF16, FP8 E4M3, FP8 E5M2, and E8M0 block scale, and an IEEE P3109 v3.2.0 cross-walk that maps each pack to its corresponding standards-track configured format. Each pack is a self-contained JSON document with a SHA-256 fingerprint, a shared row schema, and an anchor vector that encodes 3.0 -- the identity phi^2 + 1/phi^2 = 3 -- as a cross-pack sanity check. Packs are cross-validated against ml_dtypes 0.5.4 (Google/JAX); any divergence is documented explicitly and interpreted as a spec-permitted interpretation gap rather than hidden. The work is framed as registry filling: it does not propose new formats, make model-accuracy claims, or assert superiority over any vendor's implementation. All artifacts are publicly available at this https URL under an open license.

19 pages. v3: retitled Golden Ruler (count removed from title; it is a catalog invariant). Catalog now 109 formats in 12 clusters (83/13 at v2): adds the TNF, BNF and GF-T ladders with conformance vectors, folds decimal into IEEE. Sec. 6 corrected: tt-trinity-corona is not a post-silicon oracle; no die was fabricated. Source: this http URL. ORCID 0009-0008-4294-6159

Hardware Architecture (cs.AR); Artificial Intelligence (cs.AI); Mathematical Software (cs.MS); Performance (cs.PF); Numerical Analysis (math.NA)

G.1.0; D.3.4; B.2.4

arXiv:2606.09686 [cs.AR]

(or arXiv:2606.09686v3 [cs.AR] for this version)

Full text under the publisher's open licence · 287 words

Related coverage