qwen35

A native .NET LLM inference engine for GGUF models. TensorSharp provides a console application, a web-based chatbot interface, and Ollama/OpenAI-compatible HTTP APIs for programmatic access. It supports Windows/MacOS/iOS/Linux with full GPU capability

The local inference server for coding agents. Pure Rust, one binary, Apple Silicon + NVIDIA. Anthropic + OpenAI APIs; the KV cache survives across turns, so turn 20 starts as fast as turn 2. OPD training on the same runtime.

Minimal C++ LLM Qwen/Gemma inference engine

GRPO training that runs until you stop it on a single RTX4090 with vllm 0.26.0 (Linux Only).

Ace-Step-1.5XL-Ollama is a modified Ace-Step 1.5XL comfyui workflow which allows you to enter a simple song description, hit run and the workflow takes care of the rest. It uses a custom node (provided) which allows an ollama generate nodes to control the `bpm`, `timesignature`, and `keyscale`. This is a wip.

Qwen3.5 in your browser: wllama WASM build + GGUF chat playground with live Vercel demo.

run qwen 3.5 models on CPU (300KB binary)

Experiments with mixed INT4/INT8 storage for Qwen3.5 recurrent state.