Metraipolis › archive › dc_8a29cff5aec0

Transformer Accelerator (TFA): A Macro-Op INT8 Hardware Chip for Transformer Inference and Machine Translation — arXiv:2608.23582

Source
arXiv (en)
Original
https://arxiv.org/abs/2608.23582
License (as archived)
CC0-1.0
Archived
2026-10-18T06:20:26.280Z
SHA-256 of the archived text
50e55120819e668257b8eb53cc3dc999e9e4cafdc9b849082b9e3906483be55a
Stable URL of this page
https://metraipolis.com/documento/dc_8a29cff5aec0

How to cite

Transformer Accelerator (TFA): A Macro-Op INT8 Hardware Chip for Transformer Inference and Machine Translation — arXiv:2608.23582 — arXiv (en), https://arxiv.org/abs/2608.23582 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_8a29cff5aec0 (Metraipolis archive dc_8a29cff5aec0, sha256 50e55120819e668257b8eb53cc3dc999e9e4cafdc9b849082b9e3906483be55a)
BibTeX
@misc{dc8a29cff5aec0,
  title = {Transformer Accelerator (TFA): A Macro-Op INT8 Hardware Chip for Transformer Inference and Machine Translation — arXiv:2608.23582},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.23582},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_8a29cff5aec0; sha256 50e55120819e668257b8eb53cc3dc999e9e4cafdc9b849082b9e3906483be55a},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.23582. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

We present the Transformer Accelerator (TFA), a synthesizable, parameterizable INT8 memory-to-memory engine for transformer inference. One time-multiplexed datapath handles prompt processing and autoregressive generation. TFA implements matrix multiplication, softmax, RMSNorm, elementwise, and copy/gather operations through eight 512-bit macro-op descriptors. Offline-compiled programs are fetched, validated, and dispatched through AXI interfaces, supporting encoder, decoder, and encoder-decoder models. The RTL combines an output-stationary multiply-accumulate array with ping-pong buffers that overlap DMA and compute, bit-exact reciprocal-square-root and divide units, key-value-cache and embedding addressing, and an abort-safe zero-padding write engine. A UVM environment byte-compares outputs against a bit-exact golden model. Across 25 tests and 34 constrained-random runs, TFA achieved zero mismatches, 100% functional coverage, and 94.96% code coverage. We compiled the t5-small encoder-decoder pipeline for English-to-French, German, and Romanian translation. On ten multilingual proverbs, TFA executed 70,320 descriptors and matched 37.9 MB of golden-model output with zero mismatches. INT8 output matched the floating-point reference token-for-token on five sentences; the rest produced valid alternative translations. Randomized-Hadamard reparameterization recovered about 11 dB of per-tensor INT8 signal-to-noise ratio across layers. The verification configuration achieved about 20x end-to-end speedup over a 22-thread CPU, while larger designs are projected to reduce energy per token by about 1000x. After RAM inference recoding, logic area fell to 2.73 mm2, and the design completed design-rule-clean synthesis and place-and-route on SkyWater sky130. TFA demonstrates end-to-end, bit-exact execution of pretrained transformers using compact hardware and compiler-managed quantization.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-26 in cs.CL).
Categorie arXiv: Hardware Architecture (cs.AR) ; Computation and Language (cs.CL); Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2608.23582 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-18T06:17:29.387Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-08-26?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).