Code Models
AI-powered coding assistants for development
Modelos de generación de código para autocompletado, revisión y refactorización
Los modelos de generación de código son grandes modelos de lenguaje entrenados o afinados específicamente con código fuente. Impulsan el autocompletado en IDE, la revisión de PR, la refactorización automatizada, la generación de pruebas y la traducción entre lenguajes. Recurres a un modelo de código — antes que a uno general de texto — cuando quieres mayor corrección en tareas de programación y salidas estructuradas (diffs, JSON) que encajan bien con las herramientas de desarrollo.
21 models available
Codestral
Mistral's code-specialized model. Optimized for code generation, completion, and understanding across 80+ languages.
Code Llama 13B Instruct
Meta's 13B Code Llama tuned for instruction following. A faster mid-size option for code generation and completion, supporting infilling for inserting code at a cursor position. Served on Replicate per call.
Code Llama 34B Instruct
Meta's 34B Code Llama tuned for instruction following. A balance of size and quality for code generation, completion, and explanation, with strong coverage of Python, JavaScript, and other common languages. Runs on Replicate per call.
Code Llama 70B Instruct
Meta's largest Code Llama, a 70B Llama-2 derivative specialized for programming and tuned to follow instructions in chat form. Handles code generation, completion, and explanation across common languages. Served on Replicate as a per-call endpoint.
Code Llama 7B Instruct
Meta's smallest Code Llama at 7B parameters, tuned for instruction following. The cheapest and fastest member of the family for quick code generation, completion, and infilling. Served on Replicate per call.
CodeGen 350M Mono
350M autoregressive code generation model from Salesforce, the smallest of the original CodeGen family. The mono variant was further trained on Python so it is well suited for short Python completions and program synthesis from a natural-language or code prompt.
DeepSeek Coder 1.3B Instruct
1.3B instruction-tuned code model from DeepSeek, trained on 2 trillion tokens of code and natural language across 87 languages with a 16k context window. One of the strongest tiny coders for its size, handling generation, completion and short coding instructions.
DeepSeek Coder 33B Instruct (GGUF)
Quantized GGUF build of DeepSeek's 33B code model, trained on roughly 2T tokens that are about 87 percent code. Designed for repository-level completion and project-aware generation thanks to a 16k context window. Runs on Replicate as a per-call endpoint.
DeepSeek Coder V2
DeepSeek's specialized coding model. Excellent at code generation, debugging, and explanation.
Granite Code 20B
IBM Granite 20B Code Instruct. Larger Granite code model balancing quality and inference cost for enterprise CI/CD code-review automation.
Granite Code 8B
IBM Granite 8B Code Instruct. Trained on permissively-licensed code, strong on multi-language code completion and instruction-following.
Grok Build 0.1
xAI's Grok coding-focused model. Tuned for code generation and software development tasks with a 256k token context window for working over large codebases.
Magicoder S CL 7B
UIUC Magicoder S CL 7B. CodeLlama-7B fine-tuned with OSS-Instruct synthetic data. Strong HumanEval Plus and MBPP Plus performance per parameter.
Phind CodeLlama 34B v2
Phind CodeLlama 34B v2. Highly tuned CodeLlama variant focused on retrieval-augmented developer assistant workflows.
Qwen2.5-Coder 32B Instruct
Alibaba's largest open Qwen2.5-Coder model. Trained on a code-heavy corpus, it matches or beats much larger general models on code generation and repair benchmarks like HumanEval and MBPP, and supports over 40 programming languages with fill-in-the-middle completion.
Qwen2.5-Coder 7B Instruct
The 7B instruct member of Alibaba's Qwen2.5-Coder family. A lighter, faster option for code completion, generation, and bug fixing across 40+ languages, with a 128k context and fill-in-the-middle support. Good price-to-quality balance for everyday coding tasks.
Replit Code v1 3B
Replit's 3B code-completion model, trained on a permissively licensed code subset of the Stack across 20 programming languages. Built for low-latency autocomplete rather than chat. Served on Replicate per call.
Replit Code v1.5 3B
3B code completion model from Replit trained on roughly 1 trillion tokens of permissively licensed code across 30 programming languages, with a 4k context window. Designed for autocomplete-style code generation and fill-in-the-middle.
Stable Code Instruct 3B
Instruction-tuned 3B code model from Stability AI, fine-tuned from stable-code-3b for chat-style coding tasks. Handles code generation, explanation and fix-up across multiple languages and was competitive with larger code models on benchmarks at release.
StarCoder2 15B
BigCode StarCoder2 15B code-generation flagship. Trained on 4T tokens of Stack v2 data with grouped-query attention and 16k context.
WizardCoder 33B
WizardLM WizardCoder 33B v1.1. Evol-Instruct fine-tune of DeepSeek-Coder-33B with strong code-generation benchmark performance.
Top code models picks
Hand-picked across four common criteria — resolved against the live catalog so the picks track price and performance changes.
Mistral's code-specialized model. Optimized for code generation, completion, and understanding across 80+ languages.
Learn more350M autoregressive code generation model from Salesforce, the smallest of the original CodeGen family. The mono variant was further trained on Python so it is well suited for short Python completions and program synthesis from a natural-language or code prompt.
Learn moreMistral's code-specialized model. Optimized for code generation, completion, and understanding across 80+ languages.
Learn moreMistral's code-specialized model. Optimized for code generation, completion, and understanding across 80+ languages.
Learn moreLa tarificación en generación de código sigue el mismo modelo por token que el texto general. Los modelos punteros de código (GPT-5 Codex, Claude 4.6 Sonnet, Codestral) cuestan 1-10 € por millón de tokens de entrada; los niveles económicos (Codestral Mamba, DeepSeek Coder, Qwen Coder) cuestan 0,05-0,50 € por millón. Una sola petición de autocompletado en IDE rara vez supera unos pocos miles de tokens de entrada, así que el coste por llamada son fracciones de céntimo. Las facturas crecen cuando lanzas agentes que se vuelven a invocar docenas de veces por tarea.
El triángulo de compromiso es corrección, velocidad y contexto. Los punteros resuelven problemas más duros y siguen las convenciones del proyecto con más fiabilidad pero responden a 30-80 tokens/segundo, lo cual se siente lento dentro de un bucle de autocompletado ajustado. Los modelos económicos rápidos (Codestral Mamba, GPT-5 Mini) streaman a más de 200 tokens/segundo y se sienten nativos en el editor. Para tareas por lotes (refactorizar todo un repo, generar pruebas para cincuenta archivos), la corrección puntero gana. Para bucles de autocompletado ajustados, gana el nivel rápido.
Cuidado con el contexto entre archivos: la mayoría de los bucles de autocompletado solo envían el archivo actual. Para refactorización realmente consciente del codebase, necesitas una capa de recuperación que tire los archivos relacionados al prompt. Herramientas como Cursor y Continue lo hacen automáticamente; si lo desarrollas tú, embebee primero el codebase y recupera los 5-10 archivos más relevantes por petición.
Cuidado con la contaminación de licencias: unos pocos modelos de código de pesos abiertos se entrenaron solo con código de licencias permisivas; otros barrieron código GPL con términos de redistribución poco claros. Si vas a publicar código generado en un producto de código cerrado, prefiere modelos comerciales con garantías explícitas de licencia de código.
Las selecciones principales arriba cubren el puntero más correcto, el caballo de batalla más barato, el modelo con el contexto más largo y la opción de autocompletado más rápida.
Popular use cases
Common patterns built with code models on Railwail.
Frequently asked questions
Start Building with AI
Access all models through a single API. Get free credits when you sign up — no credit card required.