Local inference that behaves like a frontier API: Qwen3.8-Flash-Next on AMD Strix Halo plus small NPU models, over the OpenAI and Anthropic APIs on one port.
- amd
- gfx1151
- gpu-inference
- hip
98,836 visible repositories in the index
697–720 of 1,085 · Most stars
Local inference that behaves like a frontier API: Qwen3.8-Flash-Next on AMD Strix Halo plus small NPU models, over the OpenAI and Anthropic APIs on one port.
DeepSeek Harness (DSH) ecosystem: curated plugins, tools, and infrastructure from dsh-external/hub and the public dsh-plugin topic.
Build, Evaluate, and Deploy GUI Agents — online RL training, standardized benchmarks, and real-device deployment in one framework.
A Docker-powered service for PDF document layout analysis. This service provides a powerful and flexible PDF analysis service. The service allows for the segmentation and classification of different parts of PDF pages, identifying the elements such as texts, titles, pictures, tables and so on.
Medical o1, Towards medical complex reasoning with LLMs
The repository host gives no description.
Scalable and memory-optimized training of diffusion models
A commercially-viable ChatGpt web application built with React. 使用React搭建的一款可商业化的ChatGpt Web应用。
Stop your AI from making things up — it proposes, deterministic tools decide, every claim checked against ground truth with evidence. Grounded facts and context survive resets. Reverse engineering is the proving ground. MCP server + CLI.
Run 284B DeepSeek-V4-Flash (<59GiB RAM: 41.1 cold pp, 9.1 N=1 tg, up to 25.3 spec tok/s) & 30B MoEs (22.4 N=1 / 111.6 spec tok/s) on 64GB CPU (0 GPUs).
Hypernetworks that adapt LLMs for specific benchmark tasks using only textual task description as the input
把英文论文读成舒服的中文:本地 PDF 论文翻译、原文对照、边读边问 AI、文献管理。Read English papers in comfortable Chinese.
HuatuoGPT, Towards Taming Language Models To Be a Doctor. (An Open Medical GPT)
实时交互数字人,可自定义形象与音色,支持音色克隆,对话延迟低至3s。Real-time voice interactive digital human, customizable appearance and voice, supporting voice cloning, with initial package delay as low as 3s.
Azure OpenAI Service Proxy. Convert OpenAI official API request to Azure OpenAI API request. Support GPT-4,Embeddings,Langchain. Adapter from OpenAI to Azure OpenAI.
An AI knowledge base/agent built with .Net 9, AntBlazor, Semantic Kernel, and Kernel Memory, supporting local offline AI large models. It can run offline without an internet connection. Supports Aspire for monitoring application data.
Lifecycle-Aware Memory for long-horizon LLM agents — 66.05% on PaperBench, 94.66% on SurveyBench, 10 peer-reviewed acceptances at FSE/ICML/TOSEM/AEI/ICoGB
A simple mesh network communications app powered by the Reticulum Network Stack.
A lightweight adapter bridges SAM with medical imaging [MedIA]
An elegent pytorch implement of transformers
OpenTelemetry Instrumentation for Agent Observability
ProtTrans is providing state of the art pretrained language models for proteins. ProtTrans was trained on thousands of GPUs from Summit and hundreds of Google TPUs using Transformers Models.
AI gateway / AI control plane / AI proxy written in Go. Unified OpenAI-compatible and Anthropic-compatible API for OpenAI, Anthropic, Gemini, Groq, xAI, Ollama, vLLM and more. A LiteLLM alternative with observability, guardrails, streaming, cost tracking, intelligent routing, sticky sessions, failover, real-time logs and usage tracking. Prod ready.
character-policy: Official Benchmark & Codebase for Multimodal 3D Character & Holistic Co-Speech Gesture Policy