# Code generation optimization

AceReason-Nemotron-14B is a mathematical and code reasoning model trained through reinforcement learning, which performs excellently in multiple mathematical and code reasoning benchmark tests.

AceReason-Nemotron-7B is a mathematical and code reasoning model trained based on reinforcement learning. It starts training from DeepSeek-R1-Distilled-Qwen-7B and performs excellently in multiple benchmark tests.

A language model fine-tuned based on Qwen3-1.7B, which improves mathematical reasoning ability through reinforcement learning technology

This is a text generation model built based on the transformers library, with an efficient parameter architecture and excellent text generation capabilities.

This is a 4-bit GPTQ quantized version based on the mistralai/Devstral-Small-2505 model, optimized for consumer-grade hardware.

Marin 8B is an open-source large language model with a scale of 8B parameters, developed based on the Llama architecture and supports English text generation tasks.

A large language model fine-tuned on a specific dataset based on the ByteDance-Seed/Seed-Coder-8B-Base model, supporting long sequence input and efficient training strategies.

Seed-Coder is an 8B-scale open-source code model family, including base, instruction, and reasoning versions. The reasoning version enhances reasoning capabilities through reinforcement learning training and supports 64K context length.

Based on andrewzh's Absolute_Zero_Reasoner-Coder-14b model, this is a version quantized using llama.cpp's imatrix, suitable for reasoning and code generation tasks.

OlympicCoder-7B is a code generation model optimized based on Qwen2.5-Coder-7B-Instruct. It uses the IQ-DynamicGate ultra-low bit quantization technology and is designed for memory-constrained environments.

Ultra-low-bit quantization (1-2 bits) model using IQ-DynamicGate technology, suitable for memory-constrained devices and edge computing scenarios

Llamacpp imatrix quantized version based on PocketDoc/Dans-PersonalityEngine-V1.2.0-24b, supporting multiple quantization options, suitable for text generation tasks.

A 3B-parameter large language model based on the Llama3.2 architecture, supporting English text generation tasks, quantized using llama.cpp with imatrix

A quantized version of Qwen2.5-Coder-14B-Instruct-abliterated, supporting multiple quantization types and suitable for different hardware conditions.

A 3-billion parameter instruction-tuned language model developed by IBM, based on Granite-3.0 architecture, supporting multilingual tasks and commercial applications

Yi-Coder-1.5B is an open-source code language model with 1.5 billion parameters, supporting 52 programming languages and featuring 128K tokens of long-context understanding capability.

An instruction-following model fine-tuned based on OpenLlama-3B, optimized for role-playing, instruction following, and code generation

Phi-2 is a small yet powerful language model developed by Microsoft, featuring 2.7 billion parameters, focusing on efficient inference and high-quality text generation.

Phi-1.5 is a 1.3 billion parameter natural language processing model focused on text generation and code generation tasks, excelling in commonsense understanding and logical reasoning.

Llama 2 is a 7B-parameter dialogue-optimized large language model developed by Meta. This version extends function calling capabilities, supporting structured JSON format responses.

A code generation model fine-tuned on the APPS dataset based on GPT-Neo-1.3B, optimized for solving programming tasks