A selective view of experimental software and technical inquiry. We report working mechanisms, measured findings, and limitations separately. Source implementations remain private unless deliberately released.
Resident Model Lab is an implemented local research platform for studying how an existing mixture-of-experts language model can operate with fewer of its components held in fast GPU memory at once. The software manages component loading and residency, then measures GPU memory, logical expert reads, response time, and output fidelity.
An experimental command-line runtime manages a bounded GPU-resident expert set. Selection v2 keeps the router's expert choices intact while a layer-partitioned cache decides which expert weights stay GPU-resident. Generated tokens were identical across all 128 tested requests.
Conceptual software workflow / Not a visual of measured data
Selection v2 always executes the experts requested by OLMoE's original router. It is not model compression and does not substitute experts. Nurowerx builds the surrounding local runtime; the foundation model is existing open work.
Expert caches are partitioned by layer, with protection for experts the current request needs, replacing a single global LRU cache. The complete expert store (about 12 GiB) remains on local storage and is loaded on demand.
Results from one small experiment with serial, nonrandomized passes and an empty application cache. Timings are subject to caching, thermal, and CPU effects. Each figure is annotated with what it does and does not show.
New 128-slot vs prior 256-slot default. Caused by halving cache capacity, not by the new algorithm alone — and mean latency was 11.4% higher.
16 engineering tasks × 2 cache conditions × 4 configurations. Not a claim of universal output equivalence.
Also 13.6% fewer logical expert reads (9.9% / 4.8% at 256 slots). Logical reads are not measured SSD I/O.
Includes bitwise-equal full next-token logits for three selected prompts, empty and warm cache.
| Cache strategy | GPU slots | Peak GPU tensors | Mean response |
|---|---|---|---|
| Prior global LRU | 256 | 3.92 GiB | 14.53 s |
| Layer-partitioned (v2) | 256 | 3.92 GiB | 13.10 s |
| Prior global LRU | 128 | 2.42 GiB | 17.28 s |
| Layer-partitioned (v2) | 128 | 2.42 GiB | 16.19 s |
All four configurations solved the same 10 of 16 tasks, including shared failures — fidelity was preserved, correctness was not improved. At equal capacity, layer partitioning reduced mean latency and logical expert reads.
Only eight novel independent tasks; no independent audit or full public code release. No host-memory savings, cheaper hardware, or commercial benefit is claimed.
The earlier v0.1 restricted adaptive selector substituted a 16-expert working set for the router's choices and degraded output. It is no longer the current approach; selection v2 was designed in response.
| v0.1 execution mode | Tasks passed |
|---|---|
| Exact execution | 5 / 8 |
| Ordinary quantization | 5 / 8 |
| Restricted adaptive (16 experts) | 2 / 8 |
Planned tests include an expanded set of independent tasks, long contexts, randomized replicates, isolation of which mechanism drives each effect, host-memory limitations, and overlapping exact transfers.
Research-stage software / Private repository / No consumer product announced
Small, non-sensitive examples of signal handling and object classification pipelines.
Reference patterns for evaluating models and composing intelligent workflows.
Interface studies for reviewing machine recommendations before action.
Interface patterns for transparent model explanations, uncertainty, evidence provenance, and human correction or override.
Company research is described here without disclosing unpublished source code or private experimental materials.