LLM Reference

LLM Reference is a decision-support directory built for engineers and technology leaders who need to choose the right large language model (LLM) and provider in today's fast-moving AI landscape. It tracks over 1,700 models from more than 130 providers and 235 research labs, with data refreshed weekly to include new releases, verified price changes, and benchmark updates. The core value proposition is simple: stop wasting time hunting through scattered sources and start shipping with confidence. Whether you are building a coding assistant, an agentic workflow, a writing tool, or a research pipeline, LLM Reference gives you a single, trustworthy place to compare models side-by-side, see who offers the cheapest pricing for frontier output, and browse curated editors' picks for specific tasks like coding, agents, writing, research, image generation, and video creation. The site is designed for fast triage — you can quickly identify the right model for your job, determine the most cost-effective provider, and get back to building. With a Pulse feed that highlights what changed this week, including new models, price cuts, and benchmark refreshes, LLM Reference keeps you informed without the noise. It is built by the Data Advantage project and updated daily, making it an essential resource for anyone who needs to stay current with the exploding LLM ecosystem.

Comprehensive Model Directory

Browse through a massive and constantly updated directory of 1,843 language models from 140 providers and 247 research labs. Search by task, provider, or specific model name to instantly find what you need. The directory includes detailed information on each model's capabilities, benchmark scores, and pricing so you can make an informed decision without visiting multiple websites.

Editors' Picks and Leaderboards

Stop guessing which model is best for your specific task. LLM Reference features expert-curated Editors' Picks across six key categories: Coding, Agents, Writing, Research, Image generation, and Video creation. Each pick includes a quality rating and detailed reasoning. Additionally, 18 leaderboards track the top models for specific audiences like developers, knowledge workers, and creatives, with subcategories like Long context, Tool use, and Translation.

Pulse Feed for Real Time Updates

The AI landscape changes weekly, and LLM Reference's Pulse feed tracks every single change. See a live count of new models added (177 this week), verified price cuts from providers (53 this week), and benchmark refreshes (368 this week). This feature ensures you never miss a critical price drop or a new frontier model that could save your team time and money.

Side by Side Model Comparison

Quickly compare any two models head to head to see differences in performance, pricing, and capabilities. This feature is designed for fast triage, allowing you to evaluate options like Claude Fable 5 versus GPT-5.5 or Claude Opus 4.8 versus Gemini 3 Pro. The comparison tool highlights key metrics and cost differences so you can ship with the right model and provider, fast.

Selecting a Default Coding Model

Engineering teams need a reliable, high performance model for daily coding tasks. LLM Reference tracks the latest SWE-bench scores and coding specific benchmarks to identify the best default. For example, Claude Fable 5 is currently the top coding pick with an 80.3% SWE-bench Pro score and 96% SWE-bench Verified, making it the clear choice for production coding. You can browse all eligible models and picks to find your team's perfect match.

Identifying the Cheapest Frontier Output

Cost is a critical factor when scaling AI applications. LLM Reference tracks the cheapest frontier output pricing across all providers. Currently, Hunyuan HY3 Preview via Tencent Cloud TI Platform is the most cost effective option at $0.260 per 1M output tokens. The site surfaces these price cuts weekly, so you can switch to a cheaper provider without sacrificing performance.

Building an Agentic Workflow

Agents require models that can handle long tool loops, self-correct, and stay on task. LLM Reference's Agents leaderboard and Editors' Picks highlight the best models for this use case. Claude Sonnet 4.6 is currently the top pick with a best generally available tau-bench score of 87.5, meaning it stays on task across complex multi step workflows. You can compare agent models side by side before building your pipeline.

Researching and Comparing Open Weights Models

For teams that need to run models locally or fine tune them, LLM Reference provides a dedicated leaderboard for open weights models. DeepSeek V4 Pro is currently the top open weights pick, and you can also find the cheapest open models for specific tasks. The site tracks over 1,700 models, making it easy to find the best open source alternative for your budget and performance requirements.

How often is LLM Reference updated?

LLM Reference is updated daily by the Data Advantage project. The data refresh includes new model releases, verified price changes from providers, and benchmark updates. The Pulse feed highlights exactly what changed each week, including the number of new models, price cuts, and benchmark refreshes.

How are the Editors' Picks determined?

Editors' Picks are curated by experts who analyze the latest benchmark scores, real world performance data, and pricing. Each pick includes a detailed reasoning section that explains why a model is recommended for a specific task, such as coding or writing. The picks are updated regularly as new models and benchmarks are released.

Can I compare two models directly on the site?

Yes, the Compare tool allows you to select any two models and view a side by side comparison of their capabilities, benchmark scores, and pricing. This feature is designed for fast triage, helping you identify the right model for your job and the most cost effective provider in seconds.

Is LLM Reference free to use?

Yes, LLM Reference is a free resource for engineers and technology leaders. There is no pricing information or subscription required to access the model directory, Editors' Picks, leaderboards, Pulse feed, or comparison tools. The site is a decision support directory built by the Data Advantage project to help the community navigate the exploding LLM ecosystem.