AutomataNexus LLCWhitepaper · May 2026
Large Language Model

Trident TCN
Dilated Causal TCN Language Model

The temporal convolutional variant of Trident on Hailo-10H. 8-layer dilated causal TCN achieving 2,352 FPS. The causal convolution architecture guarantees no future token leakage by construction (unlike the Conv2D attention approximation which requires careful masking). Preferred for production autoregressive generation due to the provable causality guarantee.

Author
Andrew Jewell Sr.
Organization
AutomataNexus LLC
Framework
AxonML (Rust)
Silicon
Hailo-10H

Abstract

Background

The temporal convolutional variant of Trident on Hailo-10H. 8-layer dilated causal TCN achieving 2,352 FPS. The causal convolution architecture guarantees no future token leakage by construction (unlike the Conv2D attention approximation which requires careful masking). Preferred for production autoregressive generation due to the provable causality guarantee.

Approach

The model was trained in AxonML (a pure-Rust deep learning framework) and compiled through the Hailo Dataflow Compiler (DFC 5.3.0) targeting Hailo-10H silicon. Post-training INT8 quantization was applied during the DFC compilation pass with production telemetry calibration data. The resulting Hailo Executable Format (HEF) binary executes on Hailo’s fixed-function dataflow architecture with deterministic latency and zero framework overhead at the edge.

Results

On production hardware (Hailo-10H (NexusWatch, FW 5.3.0)), Trident TCN achieves 2,352 FPS with 48.8 °C average die temperature (max 49.0 °C).

Conclusion

Trident TCN is production-ready as a single HEF binary deployed to edge devices with no external dependencies beyond the HailoRT vendor runtime. The model meets real-time latency requirements for its target large language model application.

ModelTrident TCN
DomainLarge Language Model
ArchitectureDilated Causal TCN Language Model
Target siliconHailo-10H
Measured onHailo-10H (NexusWatch, FW 5.3.0)
DFC compiler5.3.0
FrameworkAxonML v0.6 (pure-Rust, CUDA + CPU backends)
AuthorAndrew Jewell Sr. · ORCID 0009-0005-2158-7060
OrganizationAutomataNexus LLC · Fort Wayne, Indiana

Executive overview

The temporal convolutional variant of Trident on Hailo-10H. 8-layer dilated causal TCN achieving 2,352 FPS. The causal convolution architecture guarantees no future token leakage by construction (unlike the Conv2D attention approximation which requires careful masking). Preferred for production autoregressive generation due to the provable causality guarantee.

2,352 FPS
Throughput
HW Latency
48.8 °C
Die Temp
10H
Target

Network I/O

Input: token embeddings [1, 256, seq, 1]. Output: vocabulary logits.

Architecture

Dilated Causal TCN Language Model

8-layer dilated causal TCN, d=256, kernel=3, dilation doubling per layer (1→128). Receptive field: 765 tokens. Residual connections with batch normalization. Vocabulary projection head via 1x1 Conv2D. On Hailo-10H, achieves 2,352 FPS at 48.8 degrees C die temperature — thermally sustainable for continuous generation.

Compilation constraints

All AxonML models targeting Hailo silicon are compiled under the fixed-function dataflow constraints: no dynamic control flow, no variable-length dimensions, all activations representable in INT8 after calibration, and no operations requiring dedicated softmax hardware (replaced with ReLU gating or depthwise convolution equivalents where necessary).

Silicon performance

Measured on production hardware via hailortcli benchmark with 5-second sustained inference. Device: Hailo-10H (NexusWatch, FW 5.3.0).

MetricMeasured Value
FPS (streaming)2,351.53
Die Temperature (mean)48.75 °C
Die Temperature (min)47.74 °C
Die Temperature (max)49.00 °C
QuantizationINT8 (post-training, DFC calibration)
DFC Compiler5.3.0
HailoRT5.3.0
Measured OnHailo-10H (NexusWatch, FW 5.3.0)
Table 03-1 — Production silicon measurements, 5s sustained inference.

Deployment

Deployed as a single HEF binary. No ONNX runtime, TensorFlow Lite, or Python inference stack required at the edge.

Target siliconHailo-10H
Measured onHailo-10H (NexusWatch, FW 5.3.0)
DFC compiler5.3.0
QuantizationINT8 (post-training, production telemetry calibration)
RuntimeHailoRT (vendor runtime)
Edge platformRaspberry Pi 5 + Hailo AI HAT+ (M.2 Key M)
Deployment procedure

Copy the .hef binary to the target device. hailortcli run loads the HEF directly into the Hailo-10H dataflow engine over PCIe. Inference begins immediately with deterministic per-frame latency. No model conversion, graph optimization, or warmup phase required.

References

  1. Jewell, A. (2026). AxonML: A Pure-Rust Deep Learning Framework for Edge Inference. AutomataNexus LLC. Technical whitepaper.
  2. Hailo Technologies Ltd. (2024). Hailo Dataflow Compiler User Guide. DFC v5.3.0.
  3. Hailo Technologies Ltd. (2024). Hailo-10H Product Datasheet.

AutomataNexus LLC · Fort Wayne, Indiana · andrew.jewellsr@automatanexus.com
Andrew Jewell Sr. · ORCID 0009-0005-2158-7060
May 2026 · All rights reserved.