Marketplace

State-of-the-art coding agent trained with reinforcement learning to solve complex software engineering tasks, navigate codebases, and edit multiple files. Achieves 59% on SWE-Bench-Verified.

A 7B parameter conversational AI agent model optimized for various tasks, available in multiple GGUF quantization formats for efficient local deployment

Quantized GGUF model files for AgentCPM-Explore with imatrix weighting, optimized for local inference across various quality/size tradeoffs. Includes 25+ quantization variants from IQ1_S to Q6_K.

A specialized AI agent model optimized for security and safety tasks, available in multiple GGUF quantization formats for efficient local deployment.

An open-source agentic AI model specialized in terminal operations, code generation, and software engineering tasks. Based on Qwen3-8B and fine-tuned on curated agent datasets for shell commands and bug fixing.

Advanced AI security agent based on Qwen2.5-7B, specialized in safety analysis, threat detection, and security-aware task execution with fine-grained control capabilities.

Universal framework for integrating GitHub Copilot and AI agents into software engineering workflows. Provides specialized agent roles, token-optimized documentation retrieval, multi-project orchestration, and workflow governance across any technology stack.

A specialized AI agent model optimized for security, safety, and agentic tasks based on Qwen2.5-7B architecture. Available in multiple GGUF quantization levels for local deployment.

A 4B parameter agent foundation model specialized in long-horizon tasks with sustained deep exploration capabilities. Performs multi-source cross-validation, dynamic search strategies, and real-time verification across complex benchmarks like GAIA, HLE, and BrowserComp.

Quantized GGUF format model specialized in mathematical reasoning and problem-solving, optimized for efficient local inference with llama.cpp

Expert assistant for working with the personalUtils codebase - a Python toolkit featuring GPT-powered chat, 4 specialized AI agents, 12 utility tools, and agentic workflows.

Comprehensive Aider AI coding assistant configuration template with settings for models, API keys, git integration, output customization, and workflow automation. Covers all available configuration options for optimal Aider setup.

Expert guidance for building mobile apps with TypeScript, React Native, Expo, and modern UI patterns following best practices for performance, accessibility, and security.

A specialized Llama 3 8B model fine-tuned for function calling and tool use, optimized for generating structured API calls and handling multi-turn conversations with tool integration.

Fine-tuned Turkish Llama 8B model that performs function calling tasks in Turkish using JSON schema definitions

Fine-tuned Llama 2 7B model for function calling using OpenAI-compatible metadata format. Suitable for commercial use with multi-turn conversation support.

A 20B parameter model specialized in function calling capabilities, supporting nested calls, function chaining, parallel functions, and parameter detection across multiple quantization formats (GGUF).

A 1.2B parameter liquid neural network model specialized in function calling and tool use, optimized for conversational AI agents that need to interact with external tools and APIs.

Deploy a lightweight 135M parameter model for function calling tasks with 92% structural validity. Perfect for edge devices, mobile apps, IoT systems, and embedded applications requiring structured function invocation without cloud dependency.

A 1.2B parameter Liquid Neural Network model specialized for converting natural language queries into structured JSON function calls. Runs efficiently on low-resource hardware while achieving 97% syntax reliability.