Skip to main content
Rase AI Intelligence Newsletter

AI tools

Software and services with practical AI capabilities.

Filters and sorting
Filter by milestone dates
Announcement
First access
Commercial release

141 profiles

Models & assistants

Ling-1T

Ant Group (inclusionAI)

A language model from inclusionAI that uses a mixture-of-experts architecture, with the developer positioning it for responses without a separate reasoning mode.

Medium confidenceEvidence checked

Business, finance, sales & productivity

XuanYuan-70B

Du Xiaoman Financial

A family of language models derived from Llama 2 and further trained to handle Chinese financial text, including an instruction-trained variant.

Medium confidenceEvidence checked

Business, finance, sales & productivity

Fin-R1

Shanghai University of Finance and Economics (AIFLM Lab)

A language model designed to study reasoning on financial questions, with its training and evaluations described in the Fin-R1 research paper.

Medium confidenceEvidence checked

Business, finance, sales & productivity

FinBERT

Prosus AI

A language model adapted for financial sentiment classification that can label text as positive, negative or neutral.

Medium confidenceEvidence checked

Models & assistants

N-ATLaS

National Centre for Artificial Intelligence and Robotics (NCAIR, Nigeria)

A Nigeria-focused language model effort with a Llama-based text checkpoint supporting Hausa, Igbo, Yoruba and Nigerian English.

Medium confidenceEvidence checked

Models & assistants

Aya 101

Cohere Labs

An instruction-tuned multilingual language model from Cohere for AI researchers, with the authors reporting support for 101 languages.

Medium confidenceEvidence checked

Creative AI

Wan 2.1

Alibaba (Tongyi Wanxiang)

An Alibaba video generation model family with variants for generating video from text or images.

Medium confidenceEvidence checked

Creative AI

CogVideoX

Zhipu AI & Tsinghua University

A Zhipu AI research model family for converting text descriptions into video using a diffusion transformer.

Medium confidenceEvidence checked

Creative AI

LTX-Video

Lightricks

A Lightricks video generation model based on a diffusion transformer, designed to make generation fast enough for interactive use.

Medium confidenceEvidence checked

Creative AI

Janus-Pro

DeepSeek

A DeepSeek multimodal model family that combines visual understanding with image generation in one research framework.

Medium confidenceEvidence checked

Models & assistants

InternVL3

Shanghai AI Laboratory (OpenGVLab)

A multimodal language model family from the InternVL project designed to interpret and reason about image and text inputs.

Medium confidenceEvidence checked

Creative AI

Dia

Nari Labs

A Nari Labs speech model that generates dialogue from a transcript and can include nonverbal sounds alongside spoken lines.

Medium confidenceEvidence checked

Creative AI

IndicF5

AI4Bharat (IIT Madras)

A speech synthesis model from AI4Bharat designed to produce audio in 11 Indian languages.

Medium confidenceEvidence checked

Creative AI

Shuka v1

Sarvam AI

A Sarvam AI model that pairs its Saaras audio encoder with a language model to interpret spoken questions in Indian languages.

Medium confidenceEvidence checked

Models & assistants

Raon-1 nano

KRAFTON AI

A KRAFTON language model designed to run on a device and support Korean and English conversation and tool-related tasks.

Medium confidenceEvidence checked

Creative AI

pi-H3 fast video generation

Stanford University (with Luma AI researchers)

The method generates video examples using the design reported by the researchers.

High confidenceEvidence checked

Software development

Qwen Code

Alibaba (Qwen team)

It helps users work with code and related files from a command-line interface.

Medium confidenceEvidence checked

Models & assistants

EKA-1-26B

Soket AI (Soket Labs)

It is a language model described in its maintainer's catalog entry.

High confidenceEvidence checked

Models & assistants

DeepSeek-V3

DeepSeek

It generates text and can be accessed through the technical interfaces described by DeepSeek.

High confidenceEvidence checked

Models & assistants

EXAONE 4.0

LG AI Research

The documented model handles text generation and instruction-following tasks.

Medium confidenceEvidence checked

Models & assistants

Kanana 1.5

Kakao

It generates text responses to prompts using the instruction-tuned variant described by Kakao.

Medium confidenceEvidence checked

Models & assistants

HyperCLOVA X THINK (SEED-Think-14B)

NAVER Cloud

It is intended to generate text for reasoning and instruction-following tasks.

Medium confidenceEvidence checked

Models & assistants

RakutenAI-7B

Rakuten Group

It generates text with support for Japanese-language use cases described by the authors.

Medium confidenceEvidence checked

Models & assistants

PLaMo 2

Preferred Networks

The model generates text in the tasks and languages described by Preferred Networks.

Medium confidenceEvidence checked