# Inference-Stack Software Development

> We change the inference stack at source level and hand the changes over as they are: the patch, the orchestration scripts and the raw data are all in R8.

URL: https://vuke.top/en/solutions/software-development · Updated November 12, 2026

Software development

## The problem

Once the hardware is in place, software is usually what stalls the project: the upstream framework does not support your card, the cache library has a defect under your load, or nobody has adapted a new model type.

## Our approach

We change the inference stack at source level and hand the changes over as they are: the patch, the orchestration scripts and the raw data are all in R8 [R8](https://vuke.top/en/evidence/R8).

## Key points and sources

- Upstream-grade engineering already demonstrated: ROCm builds of vLLM / LMCache, the parallel-read patch, a GDS adaptation layer and fs:// defect isolation [R1](https://vuke.top/en/evidence/R1) [R8](https://vuke.top/en/evidence/R8)
- Fast validation of new requirements: AI video-generation pipelines (ComfyUI / LTX-Video already running) [R6](https://vuke.top/en/evidence/R6) [R7](https://vuke.top/en/evidence/R7), private inference appliances and more
- Developer resources: about 10,000 experienced software developers (the company's own figure, open to due-diligence checks and not used as evidence)
- University cooperation: a partnership with Beijing Information Science and Technology University lets student developers, on the order of ten thousand by the university's own figure, join the ROCm ecosystem through joint courses and open-source practice

## Related test reports

- [R1](https://vuke.top/en/evidence/R1) · FX100 Comprehensive LLM Inference and Training Benchmark (8 × AMD MI308X)
- [R6](https://vuke.top/en/evidence/R6) · ComfyUI + LTX-Video 2.3 Full-Model Deployment, Adaptation and Usage Report (V2)
- [R7](https://vuke.top/en/evidence/R7) · ComfyUI + LTX-Video 2.3 Model Adaptation Report (AMD MI308X)
- [R8](https://vuke.top/en/evidence/R8) · FX100 KV Cache Test Code Export and Raw Benchmark Data

## Test jointly first, decide afterwards

Every engagement starts with a gated joint test of about 10 weeks. Four gates are passed in sequence; if one is missed, the project stops there, so you never pay for a benefit that was not demonstrated.

- **G1 · Week 2 · Delivery acceptance**: All materials delivered and inspected; every array passes its power-on self-test with all 24 drives recognized.
- **G2 · Week 4 · Single-node baseline**: The single-node 8-GPU inference baseline is reproduced within ±10% of the R1 figures; a local stress test of one array reaches 83% line efficiency against the measured anchor.
- **G3 · Week 7 · Main joint-test gate**: KV tiering must deliver both a TTFT reduction of ≥25% and a throughput gain inside the measured +29–40% band; with 8 nodes sharing one array, read-back bandwidth approaches the NIC ceiling.
- **G4 · Week 10 · Stability and sign-off**: 72 hours of continuous load without interruption; single-drive and single-link fault injection stays invisible to the workload; a signed joint-test report is issued.