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. The platform empowers teams to collaborate on model selection, share insights, and align on the best tools for their specific use cases, fostering a culture of informed decision-making and efficient development.

Comprehensive Model Directory

LLM Reference provides access to a massive, searchable directory of 1,843 language models from 140 providers and 247 research labs. This directory is updated daily, ensuring your team always has the most current information on model availability, capabilities, and pricing. You can search for models by name, task, or provider, and filter results based on specific criteria like cost, performance, or recency. This feature eliminates the need to visit multiple websites or rely on outdated spreadsheets, giving your team a single source of truth for all LLM-related decisions.

Editors' Picks for Specific Tasks

The platform features curated "Editors' Picks" for six key task categories: Coding, Agents, Writing, Research, Image Generation, and Video Creation. Each pick is backed by specific benchmark scores and real-world performance data, making it easy for teams to identify the best model for their project. For example, Claude Fable 5 is highlighted as the top coding model, while FLUX.2 Dev is recommended for image generation. This feature accelerates the decision-making process by providing expert recommendations based on rigorous analysis.

Pulse Feed for Real-Time Updates

The Pulse feed is a dedicated section that tracks weekly changes in the LLM landscape, including new model releases, verified price cuts, and benchmark refreshes. Each week, the platform highlights 177 new models, 53 price cuts, and 368 benchmark refreshes on average. This feature ensures your team never misses a critical update, allowing you to adapt your strategy quickly. The Pulse feed is designed to cut through the noise, presenting only the most relevant changes that impact your model selection and deployment decisions.

Side-by-Side Model Comparison

LLM Reference enables direct, side-by-side comparisons of two or more models, displaying their performance across key benchmarks, pricing per token, and supported use cases. This feature is invaluable for teams evaluating different options for a specific task, such as comparing Claude Fable 5 against GPT-5.5 for coding. The comparison tool highlights differences in capabilities, cost, and provider reliability, helping your team make an informed, collaborative decision that balances performance with budget constraints.

Selecting the Best Model for a Coding Assistant

A development team building an AI-powered coding assistant can use LLM Reference to quickly identify the top-performing models for code generation and debugging. By browsing the "Coding" category under Editors' Picks, they can see that Claude Fable 5 achieves 80.3% on SWE-bench Pro and 96% on SWE-bench Verified, making it the clear leader. The team can then compare pricing across providers to find the most cost-effective option, ensuring their assistant delivers high-quality results without exceeding their budget.

Optimizing Cost for High-Volume Agentic Workflows

For teams deploying agentic workflows that involve thousands of API calls per day, cost optimization is critical. LLM Reference's "Frontier Pricing" feature identifies the cheapest provider for frontier-level output, currently Hunyuan HY3 Preview via Tencent Cloud TI Platform at $0.260 per 1M output tokens. Teams can use this data to route their most expensive requests to the lowest-cost provider while maintaining performance, significantly reducing operational expenses over time.

Staying Ahead of Market Changes for Research Pipelines

Research teams that rely on LLMs for data analysis, summarization, and translation can use the Pulse feed to stay informed about new model releases and benchmark updates. For instance, if a new model like DiffusionGemma 26B A4B IT is released, the team can immediately evaluate its performance on relevant benchmarks and decide whether to integrate it into their pipeline. This proactive approach ensures the research team always uses the most capable and cost-effective tools available.

Comparing Models for a New Product Launch

When a product team is preparing to launch a new AI-powered feature, they can use LLM Reference's comparison tool to evaluate multiple models side-by-side. For example, they might compare Claude Opus 4.8, GPT-5.5, and Gemini 3 Pro for a writing and summarization feature. The platform's detailed benchmark data and pricing information enable the team to align on a single recommendation, reducing decision fatigue and accelerating the launch timeline.

How often is the model directory updated?

The model directory is updated daily, with new models, price changes, and benchmark refreshes added as soon as they are verified. Each week, the Pulse feed reports an average of 177 new models, 53 price cuts, and 368 benchmark refreshes, ensuring you always have access to the most current data.

What criteria are used for the Editors' Picks?

Editors' Picks are based on a combination of benchmark scores, real-world performance, and cost-effectiveness. Each pick is tagged with specific metrics, such as SWE-bench scores for coding models or Chatbot Arena ELO for writing models. The picks are reviewed and updated regularly to reflect the latest market developments.

Can I compare models from different providers?

Yes, the side-by-side comparison tool allows you to compare any two or more models from different providers. You can view their performance on key benchmarks, pricing per token, and supported use cases, making it easy to identify the best combination of capability and cost for your specific needs.

Is there a way to track price changes over time?

The Pulse feed and Changelog sections track all verified price changes, including reductions and increases. You can see a historical record of pricing adjustments for each provider and model, helping you make informed decisions about when to switch providers or negotiate contracts.

LLM Reference is currently free to use. All features, including the model directory, Editors' Picks, Pulse feed, and comparison tools, are accessible without any subscription or payment. The platform is supported by the Data Advantage project and is updated daily to provide the most current information to the AI development community. There are no premium tiers or hidden costs, making it an accessible resource for teams of all sizes.