AIBase
EN

AI News

View More

Locally Run 100 Billion Parameter AI Model: Acrab Launches Edge AI Chip GΞLIX 1, 5nm Process Enables 650TOPS

9.6k 3 minutes ago
Locally Run 100 Billion Parameter AI Model: Acrab Launches Edge AI Chip GΞLIX 1, 5nm Process Enables 650TOPS

Zeng Guoyang, CTO of Mianbi Intelligence: From Typewriters to Large Models - The Evolution and Breakthrough of Edge AI

13.7k 2 hours ago
Zeng Guoyang, CTO of Mianbi Intelligence: From Typewriters to Large Models - The Evolution and Breakthrough of Edge AI

27B Large Model Fits into iPhone! Apple Focuses on AI Compression Tech: Volume Reduced to 1/14, Speed Increased 8 Times

15.7k 4 hours ago
27B Large Model Fits into iPhone! Apple Focuses on AI Compression Tech: Volume Reduced to 1/14, Speed Increased 8 Times

Step Astronomy to Launch the World's First AI Agent Phone from a Major Large Model Vendor

15.3k 3 days ago
Step Astronomy to Launch the World's First AI Agent Phone from a Major Large Model Vendor

AI Products

View More

Vdoo AI

Vdoo AI integrates cutting - edge AI models, enabling easy creation of professional - quality films and pictures.

Video generation
8.6k

Oxlo.ai

Oxlo.ai provides cutting - edge AI model inference APIs with flexible pricing and privacy protection.

API service
5.4k

HiAPI

A one-stop multimodal AI API platform, enabling access to cutting-edge large models such as images, videos, music, and text with a single key.

API service
6.4k

VibeVoice

Open-source cutting-edge speech AI model that supports speech recognition and text-to-speech.

Voice recognition
7k

Models

View More

Gemini 2.0 Flash

Google

Gemini 2.0 Flash

$0.7

Input tokens/M

$2.8

Output tokens/M

1k

Context Length

Claude Haiku 4.5

Anthropic

Claude Haiku 4.5

$7

Input tokens/M

$35

Output tokens/M

200

Context Length

Gemini 2.5 Flash

Google

Gemini 2.5 Flash

$2.1

Input tokens/M

$17.5

Output tokens/M

1k

Context Length

Claude Sonnet 4.5

Anthropic

Claude Sonnet 4.5

$21

Input tokens/M

$105

Output tokens/M

200

Context Length

Gemini 2.5 Flash-Lite

Google

Gemini 2.5 Flash-Lite

$0.7

Input tokens/M

$2.8

Output tokens/M

1k

Context Length

qwen3-omni-flash-realtime

Alibaba

qwen3-omni-flash-realtime

$3.9

Input tokens/M

$15.2

Output tokens/M

64

Context Length

qwen3-tts-flash-realtime

Alibaba

qwen3-tts-flash-realtime

-

Input tokens/M

-

Output tokens/M

-

Context Length

Doubao-1.5-pro-32k

Bytedance

Doubao-1.5-pro-32k

$0.8

Input tokens/M

$2

Output tokens/M

128

Context Length

Doubao-Seedance-1.0-pro

Bytedance

Doubao-Seedance-1.0-pro

-

Input tokens/M

-

Output tokens/M

-

Context Length

Hunyuan-T1-20250822

Tencent

Hunyuan-T1-20250822

$1

Input tokens/M

$4

Output tokens/M

32

Context Length

Hunyuan-T1-latest

Tencent

Hunyuan-T1-latest

$1

Input tokens/M

$4

Output tokens/M

32

Context Length

DeepSeek-V3.1

Deepseek

DeepSeek-V3.1

$4

Input tokens/M

$12

Output tokens/M

128

Context Length

Tencent Hunyuan Video Generation

Tencent

Tencent Hunyuan Video Generation

-

Input tokens/M

-

Output tokens/M

-

Context Length

gpt-oss-20b

Openai

gpt-oss-20b

$0.4

Input tokens/M

-

Output tokens/M

128

Context Length

GPT-5 mini

Openai

GPT-5 mini

$1.75

Input tokens/M

$14

Output tokens/M

400

Context Length

Claude Opus 4.1

Anthropic

Claude Opus 4.1

$105

Input tokens/M

$525

Output tokens/M

200

Context Length

GLM-4.5-Flash

Chatglm

GLM-4.5-Flash

-

Input tokens/M

-

Output tokens/M

128

Context Length

GLM-4.5-AirX

Chatglm

GLM-4.5-AirX

-

Input tokens/M

-

Output tokens/M

128

Context Length

GLM-4.5-Air

Chatglm

GLM-4.5-Air

-

Input tokens/M

-

Output tokens/M

128

Context Length

Spark X1

Iflytek

Spark X1

$2

Input tokens/M

-

Output tokens/M

-

Context Length

Kani Tts 370m

nineninesix

K

KaniTTS is a high-speed, high-fidelity text-to-speech model optimized for real-time conversational AI applications. It adopts a two-stage pipeline architecture, combining a large language model and an efficient audio codec, achieving excellent speed and audio quality. It supports multiple languages and is suitable for edge/server deployment.

Audio ProcessingTransformersTransformersMultiple Languages
nineninesix
3.5k
88

LFM2 2.6B GGUF

Mungert

L

LFM2 is a new-generation hybrid model developed by Liquid AI, designed specifically for edge AI and device-side deployment. It sets a new standard in terms of quality, speed, and memory efficiency. This model adopts an innovative hybrid Liquid architecture with multiplicative gates and short convolutions, supporting multilingual processing.

Natural Language ProcessingTransformersTransformersMultiple Languages
Mungert
396
1

LFM2 2.6B GGUF

LiquidAI

L

LFM2 is a new-generation hybrid model developed by Liquid AI, designed specifically for edge AI and device-side deployment. This model sets a new standard in terms of quality, speed, and memory efficiency, and is particularly suitable for running in resource-constrained environments.

Natural Language ProcessingGgufGgufMultiple Languages
LiquidAI
278
7

Neutts Air Q8 Gguf

neuphonic

N

NeuTTS Air is the world's first ultra-realistic, edge-side text-to-speech (TTS) language model with instant voice cloning capabilities. Built on a large language model backbone with 0.5B parameters, it brings natural voice, real-time performance, built-in security, and speaker cloning capabilities to local devices.

Audio ProcessingGgufGguf
neuphonic
993
22

Loggenix Moe 0.3B A0.1B E3 Lr7e5 B16 4090 V5.1 Finetuned GGUF

mradermacher

L

This is a statically quantized version based on the kshitijthakkar/loggenix-moe-0.3B-A0.1B model, specifically optimized for low-resource environments and edge AI applications, offering multiple quantization levels.

Natural Language ProcessingTransformersTransformersEnglish
mradermacher
149
1

LFM2 700M GGUF

LiquidAI

L

LFM2 is a new-generation hybrid model developed by Liquid AI, designed specifically for edge AI and device-side deployment, setting new standards in terms of quality, speed, and memory efficiency.

Natural Language ProcessingGgufGgufMultiple Languages
LiquidAI
661
21

LFM2 350M GGUF

LiquidAI

L

LFM2 is a new-generation hybrid model developed by Liquid AI, designed specifically for edge AI and device-side deployment, setting new standards in quality, speed, and memory efficiency.

Natural Language ProcessingGgufGgufMultiple Languages
LiquidAI
817
17

LFM2 1.2B GGUF

LiquidAI

L

LFM2 is a new-generation hybrid model developed by Liquid AI, designed specifically for edge AI and device-side deployment, setting new standards in quality, speed, and memory efficiency.

Natural Language ProcessingGgufGgufMultiple Languages
LiquidAI
723
30

LFM2 1.2B GGUF

unsloth

L

LFM2 is a new-generation hybrid model developed by Liquid AI, designed specifically for edge AI and device-side deployment, setting new standards in terms of quality, speed, and memory efficiency.

Natural Language ProcessingTransformersTransformersMultiple Languages
unsloth
246
10

LFM2 350M

unsloth

L

LFM2-350M is the second-generation hybrid model developed by Liquid AI, specifically designed for edge AI and device-side deployment. This model sets a new standard in terms of quality, speed, and memory efficiency. It has 350 million parameters, supports multiple languages, and is suitable for edge computing scenarios.

Natural Language ProcessingTransformersTransformersMultiple Languages
unsloth
352
1

LFM2 1.2B

LiquidAI

L

LFM2 is a new-generation hybrid model developed by Liquid AI, designed specifically for edge AI and device-side deployment, setting new standards in quality, speed, and memory efficiency.

Natural Language ProcessingTransformersTransformersMultiple Languages
LiquidAI
3k
131

LFM2 700M

LiquidAI

L

LFM2-700M is a new-generation hybrid model developed by Liquid AI, designed specifically for edge AI and device-side deployment, setting new standards in terms of quality, speed, and memory efficiency.

Natural Language ProcessingTransformersTransformersMultiple Languages
LiquidAI
264
50

LFM2 350M

LiquidAI

L

LFM2-350M is a hybrid model developed by Liquid AI, designed for edge AI and device-side deployment, with efficient training and inference capabilities.

Natural Language ProcessingTransformersTransformersMultiple Languages
LiquidAI
1.5k
62

Loggenix Moe 0.12B A0.08B E5 Lr5e4 B16 3060 V2 Finetuned

kshitijthakkar

L

LoggenixMoE133M is a lightweight Mixture of Experts (MoE) causal language model with a total of 133M parameters and 80M active parameters. This model is trained from scratch on a custom dataset containing root cause analysis, code generation, and inference tasks. It supports special tokens for agent capabilities and is suitable for deployment on edge devices and the construction of professional AI agents.

Natural Language ProcessingTransformersTransformers
kshitijthakkar
118
1

Goppa LogiLlama

goppa-ai

G

LogicLlama is a fine-tuned language model developed by Goppa AI, based on the 1-billion-parameter version of LLaMA, enhanced with knowledge injection and logical reasoning capabilities, suitable for edge device applications.

Natural Language ProcessingTransformersTransformers
goppa-ai
39
14

Tiny Agent A 3B

driaforall

T

Mini Agent-α is a lightweight AI agent trained on the Qwen2.5-Coder model series, specifically designed for edge devices, supporting Pythonic function calling methods.

Natural Language ProcessingGgufGgufMultiple Languages
driaforall
207
13

Megrez 3B Instruct

Infinigence

M

Megrez-3B is a large language model independently trained by Infinigence AI. Through the concept of hardware-software collaboration, it creates an edge-side intelligent solution with extremely fast inference, compact size, and easy-to-use features. This model compresses the capabilities of a 14B model into a 3B model and achieves excellent performance on mainstream leaderboards.

Natural Language ProcessingSafetensorsSafetensorsMultiple Languages
Infinigence
874
32

Glm Edge 4b Chat

zai-org

G

GLM-Edge-4B-Chat is a Chinese-English bilingual dialogue model with 4 billion parameters developed by Zhipu AI. It is specifically optimized for edge devices and supports efficient text generation tasks. This model is based on the Transformer architecture, featuring lightweight design and high efficiency, making it suitable for deployment in resource-constrained environments.

Natural Language ProcessingSafetensorsSafetensors
zai-org
1.2k
12

Flux Controlnet Hed Diffusers

XLabs-AI

F

A ControlNet model based on HED edge detection developed by Xlabs AI for precise control of image generation

Computer VisionDiffusersDiffusersEnglish
XLabs-AI
196
6

MCP

View More
UnrealGenAISupport

UnrealGenAISupport

A generative AI support plugin for Unreal Engine that integrates multiple cutting - edge LLM/GenAI model APIs, providing AI integration layer support in game development, including chat and structured output functions for models such as OpenAI, Claude, and Deepseek, and supporting the Model Control Protocol (MCP) to achieve advanced functions such as scene object control and blueprint generation.

cpp
19k
3.0points
opencv-mcp-server

Opencv Mcp Server

The OpenCV MCP Server is a Python - based computer vision service that provides OpenCV's image and video processing capabilities through the Model Context Protocol (MCP). It offers a full set of computer vision tools from basic image processing to advanced object detection for AI assistants and language models, including image processing, edge detection, face recognition, video analysis, and real - time object tracking.

python
14.4k
2.5points
polydev

Polydev

Polydev is a multi - model AI programming assistant that simultaneously calls multiple cutting - edge AI models such as GPT - 5, Claude, Gemini, and Grok through a single MCP server, providing a comprehensive perspective for developers' programming agents. It supports one - click installation and can be integrated into mainstream IDEs such as Claude Code and Cursor, significantly improving the code problem resolution rate and reducing costs.

typescript
5.7k
2.0points
edgedelta-mcp-server

Edgedelta Mcp Server

The Edge Delta MCP Server is a server based on the Model Context Protocol, used for seamless integration with the Edge Delta API, enabling developers to build automation tools and AI applications to analyze observational data.

go
9.4k
2.0points