Quant & research

Reproducible and auditable AI inference evidence.

Deterministic computation, reproducible outputs, cryptographic receipts, audit trails, efficient numerical execution, and advanced transformer research - for people who need methods, not slogans.

LuxiEdge's deterministic numeric workload has also undergone independent TestFort evaluation, including H100 throughput, sustained API load, and CPU/GPU hash-matching tests.

Core capabilities

Internal milestone - independent validation next

Faithful Llama 3.1 resident inference

Internal Llama 3.1 resident-inference milestone. Active-batch output matched the serial BF16 path on tested correctness gates. Independent validation of performance and energy is the next step.

Resident
Weights and KV state
GPU-resident path
BF16
Tensor Core path
H100 NVL
Exact
Batch-to-serial agreement on tested gates
Primary p32 and p128 gates
Internal
Independent validation of performance and energy is next

Correctness gates passed internally; independent reproduction is on the validation roadmap. (Internal benchmark, not an MLPerf submission.)

TRADE residual stacks

Device-resident GPU stacks for multi-layer residual compute. Public 7B-class path: 28 layers × h=3584 on H100 NVL with multi-run throughput and sustain-only J/token.

AUDIT & receipts

Where the product lane demands it: cryptographic receipts, free-ride residual checks, and null-space behavior under load - not just marketing “determinism.”

WNSM free-ride

Null-space payload bus under real CUDA load. Public pack: free-ride vs side-channel H2D, null residual ~1e-8 class, single-layer null-inject drift 0 in the stack test path.

Long-context O(N) memory

Waller streaming state scales ~O(N) in memory vs dense scores ~O(N²). Public ladder through 32k (and analytical 131k memory reduction).

Convert + serve path

HF → Luxi native convert for 7B-class weights; serve and TRADE examples for operator evaluation. Implementation lives in the engineering repo; proof is public on LuxiDemo.

Deterministic math demo

Standalone binary: JSON expressions in, results + SHA256 out. Useful for integration smoke tests - secondary to the inference-energy thesis on this site.

Where we win, and where we are still measuring

AxisStatusPublic proof
Faithful CUDA correctness Full 28-layer Qwen2-7B on H100; CPU/CUDA and HF/CUDA greedy-token match rate 1.0 under the current acceptance protocol Correctness results →
Internal prefill evidence (absolute and matched, separate) Internal absolute prefill: ~44,860 positions per second at ~0.01532 board J/position (Qwen2-7B-class, seq 128, batch 72, dual-GEMM, no vLLM arm). Internal matched comparison: ~1.18x the tested vLLM at ~12% lower board J/position at batch 16, throughput variation below 1% across runs. Internal, prefill-stage evidence, each reported separately and separately from the independent TESTfort baseline; independent reproduction is the next step. Internal prefill evidence →
Independent prefill baseline (TESTfort, 2026-07-23) Independently verified 3.10% lower board J/position vs default vLLM at 80.60% of default vLLM throughput. TESTfort third-party measured, with the validation report published. Preserved as dated independent evidence with the complete tradeoff published. Independent baseline →
Internal TRADE research (vs Hugging Face FP16) Internal TRADE research path versus the stated Hugging Face Transformers FP16 reference - separate protocol, not a production-peer comparison. Measured numbers are kept in the research archive on Proof. TRADE benchmark →
Earlier 7B-class TRADE sustain ladder @ seq≥128 Separate public protocol h100-7b-class-TRADE
Short-sequence throughput vs Flash Complete head-to-head published, both sides h100-stack12-H2H
Free-ride / AUDIT under load Differentiated thesis h100-WNSM-free-ride
Long-ctx memory scaling O(N) vs O(N²) h100-LONGCTX-scaling
Two purpose-built execution paths. LuxiEdge's faithful CUDA path runs the complete Qwen2-7B model on NVIDIA H100 hardware and has passed the current CPU and Hugging Face token-agreement tests. Separately, LuxiEdge's TRADE research path evaluates new approaches to transformer energy efficiency, throughput, memory scaling, and data movement. Published energy results apply to that research path under the stated benchmark protocol.
TRADE scope. The public 7B-class TRADE path maps model-scale architecture into specialized TRADE kernels. Use those public measurements for energy and stack evaluation, not as a drop-in Llama chat-quality or faithful-inference claim. The benchmark numbers from one lane must not be presented as results from the other.