AIBase
EN

AI News

View More

Programming Outperforms GPT-5.5! MiniMax Launches New M3 Large Model with Three Unique Open-Source Capabilities Globally

20.7k 9 minutes ago
Programming Outperforms GPT-5.5! MiniMax Launches New M3 Large Model with Three Unique Open-Source Capabilities Globally

Exceeding GPT-5.5! Domestic AI Large Model MiniMax M3 Officially Released

17.2k 1 hours ago
Exceeding GPT-5.5! Domestic AI Large Model MiniMax M3 Officially Released

MiniMax Launches M3 Large Model: Pioneering the MSA Architecture and Supporting 1M Context, Fully Open-Source to Compete with Overseas Flagships

27.8k 9 minutes ago
MiniMax Launches M3 Large Model: Pioneering the MSA Architecture and Supporting 1M Context, Fully Open-Source to Compete with Overseas Flagships

Beat GPT-5.5! Claude Opus 4.8 Launches, Price Drops by Two-Thirds

15.4k 2 hours ago
Beat GPT-5.5! Claude Opus 4.8 Launches, Price Drops by Two-Thirds

AI Products

View More

DeepCoder

An open-source 14B parameter programming model with efficient code reasoning capabilities.

Code assistant
9.7k

Claude 3.5 Haiku

A model with advanced programming, tool usage, and reasoning capabilities.

Chatbot
10.5k

DeepSeek-V2.5

An AI model integrating general and programming capabilities.

Machine Learning
9.9k

Mistral-Large-Instruct-2407

Advanced large language model with reasoning and programming capabilities.

AI model
9.3k

Models

View More

Grok 4 Fast

Xai

Grok 4 Fast

$1.4

Input tokens/M

$3.5

Output tokens/M

2k

Context Length

o3-mini

Openai

o3-mini

$7.7

Input tokens/M

$30.8

Output tokens/M

200

Context Length

GPT-5 Codex

Openai

GPT-5 Codex

-

Input tokens/M

-

Output tokens/M

-

Context Length

Claude 3 Opus

Anthropic

Claude 3 Opus

$105

Input tokens/M

$525

Output tokens/M

200

Context Length

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

qwen3-vl-235b-a22b-thinking

Alibaba

qwen3-vl-235b-a22b-thinking

$2

Input tokens/M

$20

Output tokens/M

-

Context Length

qwen3-coder-plus

Alibaba

qwen3-coder-plus

$4

Input tokens/M

$16

Output tokens/M

1k

Context Length

wan2.5-i2v-preview

Alibaba

wan2.5-i2v-preview

-

Input tokens/M

-

Output tokens/M

-

Context Length

qwen3-omni-flash-realtime

Alibaba

qwen3-omni-flash-realtime

$3.9

Input tokens/M

$15.2

Output tokens/M

64

Context Length

Kimi-K2

Moonshot

Kimi-K2

$4

Input tokens/M

$16

Output tokens/M

256

Context Length

Doubao-Seed-1.6

Bytedance

Doubao-Seed-1.6

$0.8

Input tokens/M

$2

Output tokens/M

256

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

Qianfan-VL-8B

Baidu

Qianfan-VL-8B

-

Input tokens/M

-

Output tokens/M

32

Context Length

Grok Code Fast 1

Xai

Grok Code Fast 1

$1.4

Input tokens/M

$10.5

Output tokens/M

256

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

Qwen3-30B-A3B-Instruct-2507

Alibaba

Qwen3-30B-A3B-Instruct-2507

$0.75

Input tokens/M

-

Output tokens/M

256

Context Length

Rnj 1 Instruct

EssentialAI

R

Rnj-1-instruct is an 8-billion-parameter open-weight dense model trained from scratch by Essential AI. It is optimized for code, STEM fields, agent capabilities, and tool invocation. It performs excellently in multiple programming languages and also demonstrates strong capabilities in mathematics and science, comparable to the most advanced open-weight models.

Natural Language ProcessingTransformersTransformers
EssentialAI
441.3k
142

Rnj 1

EssentialAI

R

Rnj-1 is a series of 8 billion parameter, open-weight dense models trained from scratch by Essential AI. This model is optimized for the code and STEM fields, excelling in programming, mathematical reasoning, agent tasks, and tool invocation, with capabilities comparable to the most advanced open-weight models.

Natural Language ProcessingTransformersTransformers
EssentialAI
121.5k
53

Qwen3 Yoyo V3 54B A3B Thinking TOTAL RECALL

DavidAU

Q

This is a mixture-of-experts model based on Qwen3-Coder-30B-A3B-Instruct, with 54 billion parameters and a context length of 1 million. The model has powerful programming capabilities and general scenario processing capabilities through three-step merging and Brainstorm 40X optimization. It is especially integrated with a thinking module, which can perform in-depth reasoning before answering.

Natural Language ProcessingTransformersTransformersMultiple Languages
DavidAU
107
2

Instinct GGUF

QuantFactory

I

Instinct is a code editing model based on Qwen2.5-Coder-7B. It has been fine-tuned on real-world code editing datasets and can intelligently predict the developer's next operation to maintain a smooth programming state. This GGUF quantized version is created by QuantFactory and provides efficient local inference capabilities.

Natural Language ProcessingTransformersTransformers
QuantFactory
1.2k
3

Qwen3 Coder 30B A3B Instruct GPTQ Int8

QuantTrio

Q

Based on the GPTQ quantization version of Qwen3-Coder-30B-A3B-Instruct, it has excellent text generation capabilities, is specifically optimized for code generation and programming tasks, and supports long context processing.

Natural Language ProcessingTransformersTransformers
QuantTrio
595
1

Qwen3 Coder 30B A3B Instruct MLX 8bit

lmstudio-community

Q

Qwen3-Coder-30B-A3B-Instruct is a large code generation model developed by the Qwen team, specifically optimized for programming tasks. This model has undergone 8-bit quantization processing by the MLX team and is specially optimized for Apple chips, with efficient code generation capabilities.

Natural Language ProcessingTransformersTransformers
lmstudio-community
126.2k
1

Granite 8b Code Instruct 4k GGUF

Mungert

G

Granite-8B-Code-Instruct-4K is an 8-billion parameter code instruction model developed by IBM Research. It is fine-tuned on various instruction data based on Granite-8B-Code-Base-4K. This model enhances the capabilities of instruction following, logical reasoning, and problem-solving, and is specifically designed for building coding assistants, supporting multiple programming languages.

Natural Language ProcessingTransformersTransformersOther
Mungert
2.2k
1

Blitzar Coder 4B F.1 GGUF

prithivMLmods

B

Blitzar-Coder-4B-F.1 is an efficient multilingual coding model fine-tuned based on Qwen3-4B, supporting more than 10 programming languages and having excellent code generation, debugging, and reasoning capabilities.

Natural Language ProcessingTransformersTransformersMultiple Languages
prithivMLmods
267
1

Kanana 1.5 2.1b Instruct 2505

kakaocorp

K

Kanana 1.5 is a bilingual large language model developed by Kakao, with significant improvements in programming, mathematics, and function calling capabilities. It supports a 32K token context length and can process ultra-long texts up to 128K tokens using YaRN extension technology.

Natural Language ProcessingTransformersTransformersMultiple Languages
kakaocorp
1.4k
13

Qwen3 4B NexusPrime

ZeroXClem

Q

A high-performance, multi-domain AI model built using MergeKit's Model Stock fusion technology, integrating multiple fine-tuned Qwen3-4B models, demonstrating exceptional reasoning, programming, and multi-step problem-solving capabilities in structured output and technical application scenarios.

Natural Language ProcessingTransformersTransformersEnglish
ZeroXClem
24
3

Phi 4 Reasoning Plus

unsloth

P

Phi-4 Reasoning Enhanced is a 14-billion parameter open-source reasoning model developed by Microsoft Research, optimized through supervised fine-tuning and reinforcement learning, focusing on advanced reasoning capabilities in mathematics, science, and programming fields.

Natural Language ProcessingTransformersTransformersMultiple Languages
unsloth
189
2

Phi 4 Reasoning Plus GGUF

unsloth

P

Phi-4-reasoning-plus is an open-source reasoning model developed by Microsoft Research, focusing on advanced reasoning capabilities in mathematics, science, and programming fields.

Natural Language ProcessingGgufGgufMultiple Languages
unsloth
109.6k
47

Qwen3 32B GGUF

lmstudio-community

Q

Qwen3-32B is a large language model developed by the Qwen team, supporting a context length of 131,072 tokens with strong capabilities in mathematics, programming, and commonsense reasoning.

Natural Language ProcessingGgufGguf
lmstudio-community
56.7k
7

Seed Coder 8B Reasoning

ByteDance-Seed

S

Seed-Coder-8B-Reasoning is an 8B-scale open-source code model with enhanced reasoning capabilities through reinforcement learning, supporting a context length of 65,536 and excelling in programming tasks.

Natural Language ProcessingTransformersTransformers
ByteDance-Seed
4.6k
102

DNA R1

dnotitia

D

DNA-R1 is a Korean-specialized reasoning model optimized from Microsoft's Phi-4, significantly enhancing Korean reasoning capabilities through reinforcement learning, excelling in mathematics, programming, and general reasoning tasks.

Natural Language ProcessingTransformersTransformersMultiple Languages
dnotitia
1.9k
35

Kanana 1.5 8b Base

kakaocorp

K

Kanana 1.5 is a bilingual large language model developed by Kakao Corporation, supporting English and Korean, with significant improvements in programming, mathematics, and function calling capabilities, natively supporting a 32K tokens context length

Natural Language ProcessingTransformersTransformersMultiple Languages
kakaocorp
432
7

R1 Distill Qwen 14B Roblox Luau

TorpedoSoftware

R

A dedicated model for Roblox development fine-tuned based on DeepSeek-R1-Distill-Qwen-14B, integrating knowledge of the Roblox platform and expertise in the Luau programming language, providing professional code generation and problem-solving capabilities for Roblox developers.

Natural Language ProcessingSafetensorsSafetensorsEnglish
TorpedoSoftware
1.4k
1

TechxGenus_ _CursorCore QW2.5 1.5B Gguf

RichardErkhov

T

CursorCore is a series of open-source models designed specifically for AI-assisted programming, aiming to support functions such as automatic editing and inline chat, and replicate the core capabilities of closed-source AI-assisted programming tools like Cursor. This model is fine-tuned based on Qwen2.5-Coder-1.5B, and achieves programming assistance functions by aligning programming instruction generation data.

Natural Language ProcessingTransformersTransformersOther
RichardErkhov
1.7k
1

FuxiTranyu 8B

TJUNLP

F

FuxiTranyu-8B is an open-source multilingual large language model trained from scratch. It is trained on 600 billion tokens, focusing on multilingual capabilities and supporting 43 natural languages and 16 programming languages.

Natural Language ProcessingTransformersTransformers
TJUNLP
275
5

Yi 1.5 34B Chat 16K

01-ai

Y

Yi-1.5 is an upgraded version of the Yi model, demonstrating superior performance in programming, mathematics, reasoning, and instruction-following capabilities.

Natural Language ProcessingTransformersTransformers
01-ai
807
27

MCP

View More
ssh-hub

Ssh Hub

SSH-Hub is an MCP server that provides AI programming agents with full capabilities to access remote machines via SSH, supporting simultaneous connections to multiple servers, remote file operations, Shell command execution, and directory synchronization functions.

rust
5.1k
2.5points
mcp

Mcp

The OpenFeature MCP Server is a local tool that connects AI programming assistants to OpenFeature capabilities through a standardized protocol, providing SDK installation guidance and feature flag evaluation capabilities, and supporting multiple AI development environments.

typescript
10.2k
2.5points
coder_db

Coder_db

A memory system that enhances AI programming capabilities through multi-type database integration, including vector databases, SQLite, and knowledge graphs, for storing and retrieving code patterns, algorithms, and relationships between programming concepts.

python
10.7k
2.5points
mcp-neuralmemory

Mcp Neuralmemory

MCP-KG-Memory is an MCP server with a long-term memory layer based on a knowledge graph, aiming to solve the context forgetting problem of AI programming assistants. It persistently stores project goals, constraints, strategies, and user preferences through Neo4j, enabling semantic retrieval and active learning, and endowing the AI assistant with continuous memory and context awareness capabilities.

python
7.7k
2.5points
Cursor IDE

Cursor IDE

Cursor MCP is a tool that enables seamless integration of Claude AI with desktop applications. It bridges AI capabilities with development workflows through the Cursor IDE, providing real-time programming assistance and context-aware code suggestions.

typescript
12.2k
2.5points
instagit

Instagit

Instagit is an MCP server that enables AI programming assistants to instantly analyze and understand the source code of any GitHub repository, providing accurate integration, debugging, and code analysis capabilities based on real code rather than guesswork or outdated documentation.

typescript
8.1k
2.5points
Flux Studio

Flux Studio

MCP Flux Studio is a powerful model context protocol server that integrates Flux's advanced image generation capabilities into AI programming assistants and supports Cursor and Windsurf IDEs.

python
11.9k
2.5points
droid-mcp-rs

Droid Mcp Rs

A high-performance MCP server based on Rust that encapsulates the Factory.ai Droid CLI tool, providing AI-assisted programming capabilities for MCP clients such as Claude Code. It supports full Droid parameter mapping, session management, custom models, and fine-grained permission control.

rust
9.1k
2.5points
C# Language Features

C# Language Features

Bifrost is a VSCode extension that provides an MCP server, exposing VSCode's development tools and language features to AI tools to enhance the capabilities of AI programming assistants.

typescript
10.3k
2.5points
re-stack-mcp

Re Stack Mcp

The Re-Stack MCP Server is a model communication protocol server that connects Stack Overflow and LLMs, aiming to enhance the programming assistance capabilities of LLMs by accessing Stack Overflow content in real-time and repair the knowledge feedback loop.

typescript
9.5k
2.5points
demicode

Demicode

Awesome MCP Servers is a community - maintained curated collection of MCP servers that supports multiple programming languages and service types, aiming to expand AI capabilities.

Developer tools
9.1k
2.0points
make-copilot-an-interactive-programmer

Make Copilot An Interactive Programmer

Calva MCP Server is a VS Code extension that provides AI programming assistance and REPL interaction capabilities for Clojure/ClojureScript developers, enabling AI to execute and test code in real - time and improving programming efficiency.

javascript
10.7k
2.0points
la-forge-mcp

La Forge Mcp

LA FORGE MCP is a Model Context Protocol server that provides visual perception capabilities for AI programming assistants. Through pixel-level difference comparison, computed style extraction, and CSS rule chain analysis, it helps AI accurately detect CSS/HTML rendering issues, addressing the pain point that AI can only read code but cannot truly 'see' the rendering results.

typescript
8.1k
2.0points
nabu_nisaba

Nabu_nisaba

Nabu and Nisaba are research prototype toolkits used to enhance the code understanding and development efficiency of LLM agents. Nabu, as an MCP server, provides code semantic search and structure analysis functions, supporting multiple programming languages; Nisaba provides workspace management capabilities for Claude Code through proxy injection and a TUI interface, helping agents manage context usage autonomously.

python
8.5k
2.0points
aider-mcp-server

Aider Mcp Server

An MCP server based on Aider, providing AI programming capabilities and supporting multiple models and transport protocols.

python
10.7k
2.0points
linggen-releases

Linggen Releases

Linggen is a locally-first RAG and MCP memory layer that provides cross-project shared memory, code repository navigation, and multi-tool connection capabilities for AI programming assistants. All data is processed locally by default.

Knowledge management and memory
8.2k
2.0points
mcp-servers

Mcp Servers

This project provides a series of Model Context Protocol (MCP) servers for integration with the Cursor IDE, extending the capabilities of the AI programming assistant, including file system access, memory storage, web search, etc.

javascript
8.7k
2.0points
AIBase
Empowering the future, your artificial intelligence solution think tank
English简体中文繁體中文にほんご
FirendLinks:
AI Newsletters AI ToolsMCP ServersAI NewsAIBaseLLM LeaderboardAI Ranking
© 2026AIBase
Business CooperationSite Map