LLM Reference

LLM Reference is a decision-support directory engineered for developers, engineering leads, and technology leaders who need to select the optimal large language model and provider in the rapidly evolving AI landscape. This platform tracks over 1,800 language models from more than 140 providers and 247 research labs, with data refreshed weekly to capture new releases, verified price changes, and benchmark updates. The core value proposition eliminates the inefficiency of hunting through scattered sources, enabling teams to ship with confidence. Whether you are building a coding assistant, an agentic workflow, a writing tool, or a research pipeline, LLM Reference provides a single, trustworthy location to compare models side-by-side, identify the cheapest frontier output pricing, 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, allowing you to 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 weekly changes 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

Access a searchable directory of over 1,800 language models from more than 140 providers and 247 research labs. Each model entry includes detailed specifications, benchmark scores, pricing information, and provider details. The directory is refreshed weekly to ensure you are always working with the most current data, including new releases, verified price changes, and benchmark updates. You can filter by task type, provider, pricing tier, or benchmark performance to zero in on the models that matter most for your specific use case.

Editors Picks for Task-Specific Selection

Curated by experts, the Editors Picks section provides trusted recommendations for six critical task categories: coding, agents, writing, research, image generation, and video creation. Each pick includes a performance rating, key benchmark scores, and a brief rationale explaining why the model excels for that specific task. This feature saves hours of research by pointing you directly to the best options for your workflow, whether you need a coding powerhouse like Claude Fable 5 or a photorealistic image generator like FLUX.2 Dev.

Real-Time Pulse Feed

The Pulse feed is your weekly snapshot of the model market, highlighting what changed in the last seven days. It tracks new model releases, verified price cuts from providers, and benchmark refreshes across major evaluation suites. With 177 new models, 53 price cuts, and 368 benchmark refreshes tracked weekly, the Pulse feed keeps you informed without overwhelming you with noise. You can quickly see which models are trending, which providers dropped prices, and which benchmarks were updated.

Side-by-Side Model Comparison

The Compare tool enables you to pit two models against each other across multiple dimensions, including benchmark scores, pricing, context length, and task suitability. This feature is invaluable for making informed decisions when evaluating alternatives like Claude Fable 5 versus GPT-5.5 or Claude Opus 4.8 versus Gemini 3.1 Pro Preview. The comparison view highlights key differences and helps you identify the best model for your specific requirements and budget.

Selecting a Coding Assistant Model

Engineering teams building AI-powered coding tools need a model that excels at code generation, debugging, and understanding complex programming contexts. LLM Reference helps you compare models like Claude Fable 5, which achieves 80.3 percent on SWE-bench Pro and 96 percent on SWE-bench Verified, against alternatives like GPT-5.5 or Claude Opus 4.8. You can filter by coding-specific benchmarks, compare pricing per token, and read editors picks to make a confident selection for your development pipeline.

Choosing a Model for Agentic Workflows

When building autonomous agents that interact with tools, APIs, and environments, you need a model that stays on task across long tool loops and self-corrects without prompting. LLM Reference provides curated picks for agents, highlighting Claude Sonnet 4.6 with its best generally-available tau-bench score of 87.5. You can compare agent-specific benchmarks, evaluate context handling, and identify the most cost-effective provider for your agent deployment.

Evaluating the Cheapest Frontier Output

For teams processing large volumes of text, finding the cheapest frontier model output is critical for cost management. LLM Reference tracks the cheapest frontier output pricing, currently at $0.260 per 1 million tokens via Hunyuan HY3 Preview on Tencent Cloud TI Platform. You can browse all providers, compare price cuts announced this week, and identify the most affordable option that still meets your quality and performance requirements.

Picking a Model for Research and Analysis

Knowledge workers and research teams need models that excel at analytical tasks, data interpretation, and long-form reasoning. LLM Reference features editors picks for research, highlighting Claude Fable 5 with its GDPval-AA ELO of 1932 and strong performance in finance, trading, and analytics. You can compare models across research-specific benchmarks, evaluate context window sizes, and select the best option for your data analysis and document review workflows.

How often is the model data updated?

LLM Reference updates its data weekly, with new models, verified price changes, and benchmark refreshes added every week. The Pulse feed highlights exactly what changed in the last seven days, including new model releases, price cuts, and benchmark updates. The platform is built by the Data Advantage project and updated daily to ensure you always have access to the most current information.

What criteria are used for Editors Picks?

Editors Picks are curated by experts based on a combination of benchmark performance, real-world task suitability, pricing efficiency, and provider reliability. Each pick includes a performance rating, key benchmark scores, and a rationale explaining why the model excels for that specific task. The picks are reviewed and updated regularly to reflect new model releases and benchmark updates.

Can I compare two models side by side?

Yes, the Compare tool allows you to select any two models from the directory and view them side by side across multiple dimensions. You can compare benchmark scores, pricing per token, context length, provider details, and task-specific performance. This feature is designed to help you make informed decisions quickly, whether you are evaluating coding models, agent frameworks, or creative tools.

Is LLM Reference free to use?

LLM Reference is a free resource for anyone who needs to stay current with the LLM ecosystem. There are no subscription fees or paywalls for accessing the model directory, Editors Picks, Pulse feed, or comparison tools. The platform is supported by the Data Advantage project and is available to all users who need to make informed decisions about language model selection.