AI Wiki — Knowledge Base

Finally got qwen 3.8 q4 k_M 27b runnin slick 40tkps load at 240k inf 4qnl for stable agentic coding on a 7900xtx. I have been finally able to give hermes a task and come back to results. Built a panel to manage my inference servers from hermes.

Screenshot of heavy load. I did offload vision to a spare 2060 instead of local ram. The control panel was generated with qwen3.8-27b davidau q4k_m max mtp. Made the panel from scratch over a week, ba

model-releasesr-localllama

19 Sept 2026

Search: “inference”

When Has a Bayesian Neural Network Sampled Enough? Adaptive Inference Time with Statistical Guarantees

Possibilistic Radial Transport for Approximate IM Inference

LBA-CBF: Rapidly Adaptive Safety Filters via Parallel Dynamics Inference

Looking for developer-friendly inference providers who give you enough API credits to experiment [D]

A Statistical Inference Framework for PMI Estimation and SGNS Word Embeddings

poorman inference engine for 16GB GPU and 35B moe Qwen 3.6for coding

Inference for stochastic differential equations driven by weighted sub-fractional Brownian motion using neural networks and the Euler approximation

Not All Error Yields to Scale: Where Scaling Stops in Vision-Language Inference

Beyond Token Importance: Preserving Spatial Scaffolds for Efficient Vision-Language-Action Inference

CHAIN: Calibrated LLM Forecasting via Causal-Temporal Hypergraph Inference

HARISSA: Inference-Time Self-Checks for Efficient and Safe Local Language Model Deployment

LLM Judge Validation Under Sparse Overlap: From Inference to Design

AnyStep-WAM: Budget-Aligned Distillation and Adaptive Inference for World Action Models

Elastic Selective Spectral Hybrids for Train-Once, Export-Many Budgeted Inference

GraphSelect for Budgeted Representation Selection in Multimodal Graph Inference

Seeing Parts, Reasoning about Worlds: Visual Inference under Partial Observation

Singularities of Non-negative Matrix Factorization and their application to Bayesian inference

If you are running Qwen 3.8 Flash Next on Strix Halo, use this software for inference. It's so much faster than llama.cpp especially at high context.

Query-Conditioned Prototype Adaptation for Cross-Domain Few-Shot Learning: Single-Query Inference, Controlled Comparisons, and Failure Modes

Hardware Roofline Inference Calculator

An Order-Theoretic Characterization of Consistent Inductive Inference

Paging the Experts: A Reproducible Characterization of Flash-Backed MoE Inference on iPhone

SpaFactor: Lightweight Spatial Context-Aware Gene Program Modeling for Histology-to-Transcriptomics Inference

Task-Aware Spectral Pruning: A Mixture-of-Masks Framework for Efficient LLM Inference

MicroQonv: Reshaping Convolution Tensors for Efficient Microscaling in Training and Inference

Scaling of Capability and Efficiency at Inference Time in Large Reasoning Models

Greedy Decoding Is Not Precision-Invariant: Cross-Precision Output Divergence in LLM Inference

Ultra-fast Neural Inference for Stochastic Gaussian Splatting Denoising

Zephyron: Integrated Design and Analytical Evaluation of a Solar-Assisted Mobile Manipulator for Multimodal Environmental Reconnaissance and Distributed Visual Inference

HapticWAM: Distilling Imagined Touch into a World-Action Model without Inference-Time Tactile Sensing

Leveraging Inference-Time Compute for Diffusion Models via Global Scheduling of Denoising Trajectories

Morpho-VITS: Variational Inference with Morphological Modeling for End-to-End Speech Synthesis of a Tonal Bantu Language

RLVR is a Kernel, Not a Function: Statistical Inference for pass@k Crossovers

StateMem: Single-State Residual Memory with Adaptive Inference for Vision-Language-Action Policies

Statistical Inference for Adversarial Training: Central Limit Theorems via Optimal Transport

Finally got qwen 3.8 q4 k_M 27b runnin slick 40tkps load at 240k inf 4qnl for stable agentic coding on a 7900xtx. I have been finally able to give hermes a task and come back to results. Built a panel to manage my inference servers from hermes.

SoftTri: Smooth Triangular Membership Functions for Adaptive Fuzzy Inference Systems

Beyond Static RAG: An Adaptive, Tri-Metric Routing Framework for Efficient Long-Context Inference on Commodity GPUs

Call for compute - help optimize Qwen inference speed on local hardware

From a River in Gilead to the Inference Distributions of Large Language Models: Covert Dialect Bias and Linguistic Profiling at Scale

Differentiable Mesh State Estimation via Factor Graph Inference for Deformable Object Reconstruction

From Momentary Emotion Inference to Sustained Emotion Support: Evaluating a Companion Agent in a Longitudinal Study

Evaluation of optimisation and Bayesian inference methods for reaction rates in atmospheric chemical mechanisms

How NVIDIA Groq 3 LPX Deterministic Execution Drives Power-Efficient High-Interactivity Inference on NVIDIA Vera Rubin

NeuroFlex: Lossless Element-Level ANN-SNN Co-Execution for Efficient Sparse Inference

A Brain Too Big to Carry — On-Device vs Datacenter Inference

Swarm inference AI

FastE: Readout-Triggered Token Compression for LLM Embedding Inference