SuperAPI AI Directory
Partner with us中文AI newsFavorites

All AI tools

35,738 results · Page 345 of 993

PPTAgent

AI office

PPTAgent is an agent framework for reflection-based PowerPoint generation, open-sourced by institutions such as the Institute of Software, Chinese Academy of Sciences. It supports creating, modifying, and visually reviewing editable PowerPoint presentations through coding agents (such as Claude Code, Codex, OpenCode), and supports exporting PPTX. The project includes two versions, PPTAgent (EMNLP 2025) and DeepPresenter (ACL 2026), and provides features such as MCP server support, free-form and template generation, offline mode, and context management. Users can combine multimodal vision models for slide review, and can drive text generation through models such as Atria Dawn Preview. Installation requires Linux (including WSL) or macOS, uv, npm, and dependencies such as LibreOffice.

Read more

PR1

AI developer platforms

Perception-R1 (PR1) is the official code implementation of the NeurIPS 2025 paper, proposing a scalable reinforcement learning framework that post-trains multimodal large language models (MLLMs) via Group Relative Policy Optimization (GRPO) to enhance visual perception capabilities. Its core innovations include: perceptual perplexity analysis, used to reveal the key threshold for effective reinforcement learning in perception tasks; rule-based reward shaping and GRPO optimization, enabling scalable policy learning; and significant improvements on multiple visual perception benchmarks, for example reaching 31.9% mAP on the COCO2017 validation set, becoming the first 3B-scale MLLM to achieve this performance. The project provides training and evaluation code, supports tasks such as grounding, counting, and detection, and users need to configure model and data paths themselves and separately download COCO images.

Read more

Practical Tips for Finetuning LLMs Using LoRA (Low-Rank Adaptation)

AI developer platforms

This article is Sebastian Raschka's practical tips for LoRA (Low-Rank Adaptation) fine-tuning of large language models, summarized based on hundreds of experiments. LoRA is a widely used technique for efficiently training custom LLMs and is suitable for those interested in open-source LLMs. The article shares key lessons and answers common questions, including consistency of training results, QLoRA trade-offs, optimizer selection, application of LoRA layers, hyperparameter tuning, and more, aiming to save fine-tuning time.

Read more

Prefect

AI agents

Prefect is a platform for orchestrating data, machine learning, and AI agent workflows, with its core based on the open-source Prefect framework. Developers only need to add the @flow decorator to a Python function to gain state tracking, logging, and automatic retries. It can replace fragile cron jobs, support automatic retries on failure, on-demand scaling, and task tracking, and provide a unified interface to view run status across teams and infrastructure. Prefect also provides the open-source FastMCP framework and FastMCP Gateway for building MCP servers to connect AI agents to various tools and data. It supports managed compute, your own VPC, or a hybrid deployment of both, and the company says the enterprise edition provides a 30-minute response SLA.

Read more

Prefer

AI marketing

Prefer is an answer engine optimization (AEO) platform used to track how brands are mentioned and cited in AI engines such as ChatGPT, Gemini, Google AI Overviews, Perplexity, and others, and to provide content optimization recommendations. It can run real buyer questions, record the sources cited by each engine, score brand visibility, share of voice, and sentiment, and generate recommendations for pages to publish, comparison pages, and structured data. It offers a free AEO report, generated in about 10 minutes, with no signup required. Paid plans provide continuous prompt tracking, competitor benchmarking, agent analytics, and content agents.

Read more