LLM Models · Code
Browse the world's large language models. Compare parameters, benchmarks, VRAM and more.
Code listings are aimed at programming tasks: edits, explanations, tests, repository-scale questions. The category is a hint about training focus, not a guarantee the model fits your repo in context or that it will refuse to invent an API.
Start with the window. A code model that cannot hold the package will guess the rest. Chunking is a different workflow. How to read a listing puts the window before the prose because code jobs are where people discover the cap too late. Then look at license and hosting. A code model you cannot legally run on customer source is not a code model for that customer.
Compare two code models on a diff you already understand, not on a vendor demo that completes a famous algorithm. Deciding after a comparison describes that prompt. Include a trap: a file the model must not edit. A model that "fixes" the trap failed, even if the tests it wrote are green.
This shelf overlaps development skills. A skill is a prompt that steers an agent. A model is the thing that answers. Installing a Spring Boot skill does not change the model. If the answers are weak because the model never sees the file, change the setup or the model, not the number of skills. Your first week keeps the model pinned for that reason.
Score sorts the shelf. It is not a SWE-bench run we performed. Read the benchmark block if it names a dataset you care about, and ignore bars that do not name one.
Save the diff the model was not allowed to touch, and the diff it produced. Those two files are the evaluation. If you only saved the model's explanation, you saved the advertisement. The explanation can be fluent while the forbidden file still changed.
Description: The NVIDIA Kimi-K2.7-Code NVFP4 model is the quantized version of the Moonshot AI's Kimi-K2.7-Code model, which is an auto-regressive language model that uses an optimized transformer architecture. For more information, please check here. The NVIDIA Kimi-K2.7-Code NV…
KAT-Coder-Pro V2 is the latest high-performance model in KwaiKAT’s KAT-Coder series, designed for complex enterprise-grade software engineering and SaaS integration. It builds on the agentic coding strengths of earlier versions,...
KAT-Coder-Pro V2.5 is a flagship-level Agentic Coding model that can directly hand over an entire issue or an entire business workflow to it, allowing it to autonomously locate and make...
GPT-5.3-Codex is OpenAI’s most advanced agentic coding model, combining the frontier software engineering performance of GPT-5.2-Codex with the broader reasoning and professional knowledge capabilities of GPT-5.2. It achieves state-of-the-art results...
We introduce Olmo 3, a new family of 7B and 32B models both Instruct and Think variants. Long chain-of-thought thinking improves reasoning tasks like math and coding.
Kimi K2.7 Code is a coding-focused agentic model built upon Kimi K2.6. With substantial improvements on real-world long-horizon coding tasks, it strengthens end-to-end task completion across complex software engineering workflows while improving token efficiency, reducing thinkin…
Introduction We are thrilled to introduce Seed-Coder, a powerful, transparent, and parameter-efficient family of open-source code models at the 8B scale, featuring base, instruct, and reasoning variants. Seed-Coder contributes to promote the evolution of open code models through…
Introduction We are thrilled to introduce Seed-Coder, a powerful, transparent, and parameter-efficient family of open-source code models at the 8B scale, featuring base, instruct, and reasoning variants. Seed-Coder contributes to promote the evolution of open code models through…
Introduction We are thrilled to introduce Seed-Coder, a powerful, transparent, and parameter-efficient family of open-source code models at the 8B scale, featuring base, instruct, and reasoning variants. Seed-Coder contributes to promote the evolution of open code models through…
Seed 2.0 Code is a model from ByteDance Seed optimized for agentic coding. It is suited for frontend development, multilingual programming tasks, and coding-agent workflows in tools such as Claude...
North Mini Code is an open weights research release of a 30B-A3B parameter model optimized for code generation, agentic software engineering, and terminal tasks.
North Mini Code is an open weights research release of a 30B-A3B parameter model optimized for code generation, agentic software engineering, and terminal tasks.