# Multilingual Code Generation

Devstral is an intelligent large language model specifically designed for software engineering tasks, jointly developed by Mistral AI and All Hands AI. It excels in code exploration, multi-file editing, and driving software engineering agents.

Devstral is a large language model for software engineering task agents, developed in collaboration between Mistral AI and All Hands AI. It excels at using tools to explore codebases, edit multiple files, and drive software engineering agents.

Devstral is an intelligent large language model specifically designed for software engineering tasks, developed in collaboration by Mistral AI and All Hands AI. It excels in codebase exploration, multi-file editing, and driving software engineering agents.

Devstral is an intelligent large language model developed by Mistral AI in collaboration with All Hands AI for software engineering tasks, excelling in codebase exploration, multi-file editing, and driving software engineering agents.

A code reasoning model fine-tuned on the Qwen3-4B architecture, specifically optimized for high-precision code reasoning and structured logic tasks.

This is the quantized version of the NVIDIA OpenCodeReasoning-Nemotron-32B-IOI model, processed using llama.cpp for quantization, suitable for code reasoning tasks.

GLM-4-32B-0414 is a new member of the GLM family with a parameter scale of 32 billion, comparable in performance to the GPT series, and supports local deployment.

Phi-4-mini-instruct is a lightweight open-source model focused on high-quality, reasoning-rich data, supporting a context length of 128K tokens.

OpenThaiGPT R1 32b is a 32-billion-parameter Thai reasoning model that excels in Thai mathematical, logical, and code reasoning tasks, outperforming larger-scale models.

Refact-1.6B is a large language model with 1.6B parameters, specializing in code generation and excelling across multiple programming languages.

Viper-Coder-v1.5-r999 is a large language model based on the Qwen 2.5 14B architecture, optimized for coding and reasoning tasks, with strong chain-of-thought reasoning and logical problem-solving capabilities.

Phi-3.5-MoE is a lightweight, state-of-the-art open-source model supporting multiple languages with a 128K context length, focusing on high-quality, inference-intensive data.

CodeGeeX4-ALL-9B is the latest open-source version of the CodeGeeX4 series models, continuously trained based on GLM-4-9B, significantly enhancing code generation capabilities.

The IBM Granite 8B Code Instruct model is an AI model focused on code generation and explanation, supporting multiple programming languages and excelling in benchmarks like HumanEval.

Codestral-22B-v0.1 is a 22B-parameter programming language model released by Mistral AI, supporting code generation and comprehension tasks for over 80 programming languages

The IBM Granite 34B Code Instruction Model is a large language model focused on code generation and explanation, based on a 34B parameter architecture and supporting multiple programming languages.

Phi-3-Small-8K-Instruct is a 7B-parameter lightweight open-source model focused on high-quality reasoning capabilities, supporting 8K context length, suitable for commercial and research applications in English environments.

CodeGemma is a lightweight open-source code model series based on Gemma, specializing in code generation and dialogue tasks.

Granite-20B-Code-Instruct-8K is a 20-billion-parameter code generation model, fine-tuned from Granite-20B-Code-Base-8K, specializing in code generation and instruction-following capabilities.

Granite-8B-Code-Instruct-4K is an 8-billion-parameter code instruction model, fine-tuned on various permissible instruction datasets based on Granite-8B-Code-Base-4K, enhancing its ability to follow instructions, including logical reasoning and problem-solving skills.

Granite-3B-Code-Instruct-2K is a 3-billion-parameter model fine-tuned from Granite-3B-Code-Base-2K, with enhanced instruction-following capabilities, particularly excelling in code generation and logical reasoning tasks.

Granite-3B-Code-Base-2K is a decoder-only model developed by IBM Research specifically designed for code generation tasks, featuring 3B parameters and supporting 116 programming languages.

aiXcoder-7B Base is a 7B-parameter large language model focused on code completion and generation tasks, trained on 1.2T unique code data, demonstrating excellent performance in real-world development scenarios.

CodeGemma is a series of lightweight open-source code models based on Gemma, focusing on code completion and generation tasks.

CodeGemma is a lightweight open-source code model based on Gemma, focusing on code completion and generation tasks.

CodeGemma is a lightweight open-source code model series based on Gemma, focusing on code completion, generation, and conversational tasks.

CodeGemma is a lightweight open-source code model series based on Gemma, featuring text-to-text and text-to-code decoder models, specializing in code completion and generation tasks.

StarChat2 is a 16-billion-parameter programming assistant model fine-tuned based on StarCoder2, excelling in conversation and code generation tasks

ERNIE-Code is a unified large language model connecting 116 natural languages and 6 programming languages, supporting various cross-lingual tasks.

Code Ninja is an enhanced version of the renowned model openchat/openchat-3.5-1210, trained through supervised fine-tuning on two large-scale datasets containing over 400,000 coding instructions.

StarChat-β is a language model fine-tuned based on StarCoderPlus, focusing on programming assistance tasks and supporting 80+ programming languages

Based on the StarCoder model, fine-tuned using the GPTeacher code generation dataset to optimize code generation and explanation capabilities

SQL generation model fine-tuned from Salesforce/codegen-350M-multi, specializing in JOIN operation generation

CodeGen-Multi 16B is a 16B-parameter autoregressive language model for program synthesis, supporting multilingual code generation.

CodeGen is an autoregressive language model family proposed by Salesforce for program synthesis, capable of generating code in multiple programming languages based on natural language prompts

CodeGen is a series of autoregressive language models for program synthesis, supporting code generation in multiple programming languages.