Models

A four-stage multi-agent pipeline, presented at AAAI 2026, that classifies lung adenocarcinoma nodules from CT images and reports through nodule detection, sizing, report generation and a multi-agent clinical debate.

• Models
• Papers

A two-stage benchmark comparing more than twenty optimizers for large language model pre-training at scale, first screening them broadly on a shared architecture and then testing the strongest ones for generalization across several model architectures.

• Models
• Papers

A POWER8-specific research prototype that replaces some full-matrix attention computation with a hardware AES instruction used as a fast prefilter, so only a subset of attention paths get the full computation.

• Models
• Inference

A browser-native inspector for neural network model files that opens a model and explores its computation graph, metadata, tensor shapes and weight statistics with no upload and no server.

• Models
• Debugging

A toolkit that converts a dense Gemma-3-270m model into a mixture-of-experts model, then trains, inspects and serves it on CPU through llama.cpp.

• Models
• Inference

An independent, from-scratch implementation of the Kimi K3 language model architecture described in its paper, reproducing its published parameter counts and running the same code path from a 20-million-parameter laptop model up to its 2.8-trillion-parameter configuration.

• Models

A reinforcement-learning framework for repository-level code completion that trains a retriever to select reference code snippets that actually help generation, rather than merely similar ones.

• Models
• Papers

A research toolbox for training and evaluating multimodal large language models under continual instruction tuning, where a model learns new tasks in sequence without forgetting earlier ones.

• Models
• AI evals

A Python library that profiles the execution time of individual PyTorch layers and functions on a CPU, then simulates how those same layers would run and how much energy they would use on a target hardware accelerator profile.

• Models
• Debugging

An MCP server exposing a precomputed observation database of 23 large language models across 13 architecture families, letting a coding agent ask which internal layers of different models behave alike, without running any of those models itself.

• MCP servers
• Models

A training platform that repeats a four-phase cycle to build adversarial robustness into a text-classification model without eroding the defenses learned in earlier cycles, and generates EU AI Act compliance reports.

• Models
• AI evals

A method that writes new facts directly into the n-gram lookup table some recent language models carry alongside their transformer, as a small removable overlay file, without changing any of the model's trained weights.

• Models

A quantization method for large language models that rotates weights toward a Gaussian distribution before applying optimal centroids, shrinking a model's memory footprint with close to full-precision accuracy.

• Models

The official research code for an ECCV 2026 paper on quantizing every part of a multi-agent cooperative perception system for autonomous vehicles, aimed at cutting inference and communication cost with minimal loss of perception accuracy.

• Models
• Papers

A PyTorch implementation of a published method for image classification using multi-agent reinforcement learning, where several agents move across an image and their policy is trained to classify it.

• Models
• Papers

An unofficial, paper-aligned implementation of DeepSeek's Engram architecture that injects a large, sparsely retrieved conditional memory into a transformer model without increasing inference compute.

• Models
• Papers

A from-scratch training framework for Llama 3 architecture language models, covering data preparation, pretraining, LoRA personality fine-tuning, GGUF export and a small standalone Go inference engine.

• Models

A research benchmark and training project for small vision-language models that spot factory procedure errors in first-person video by identifying exactly when a step starts and ends, aimed at real-time deviation warnings before an incident happens.

• Models
• AI evals

A Python library that runs structured natural-language tasks, such as extracting bullet points, generating question-and-answer pairs and building knowledge graphs, locally through a fine-tuned 135-million-parameter language model instead of a large general-purpose one.

• Models
• Libraries

An educational project that trains a large language model from scratch through two pipelines: a native C++ engine for data ingestion, tokenization and training, and a Python pipeline using Apple's MLX framework that scales a decoder-only GPT model to 1.6 billion parameters.

• Models
• Tutorials

A memory-optimization toolkit for fine-tuning SDXL-class and ANIMA diffusion models on 12 GB consumer GPUs, combining a custom optimizer that reduces optimizer-state memory with a per-layer scheduler that skips computation on layers whose gradients have gone flat.

• Models
• Generative media

An unofficial PyTorch implementation of the Recurrent Looped Transformer proposed by Yifan Zhang of Princeton, an encoder-decoder architecture that can share weights between encoder and decoder and vary its recurrent block size across a sequence.

• Models
• Libraries

A reinforcement-learning framework for language models that builds adaptive training environments where a model practices problem-solving and is scored with multi-dimensional, partial-credit rewards instead of a single binary right-or-wrong signal.

• Models

A framework for fine-tuning GPT-2 to generate math riddles and their solutions in multiple explanatory styles, using a dual-decoder architecture and a coherence-penalty loss that ties a generated riddle's wording to its generated answer.

• Models