What is Metatext?
Metatext is a free, open directory for discovering and comparing AI models, datasets, and tools. Metatext serves AI researchers, engineers, and builders who need to choose models, locate training or evaluation datasets, discover agent resources, or find AI roles. The directory brings together information otherwise spread across provider pricing pages, benchmark leaderboards, dataset hubs, and tool registries.
Metatext connects model performance with model cost by displaying benchmark scores alongside prices per million tokens. Model comparisons also include context windows and providers. This combination supports decisions based on cost and performance together, rather than benchmark rankings alone, while separate listings cover language models, decision models, datasets, agent tools, and jobs.
What Metatext can do
Metatext provides searchable listings and comparison views across the AI ecosystem. Metatext covers model selection, dataset discovery, agent tooling, and employment searches, with different browsing criteria for each resource type. Benchmark coverage includes LMArena, LiveBench, SWE-bench, and Aider, while model listings provide specifications, lineage, and metadata.
- Metatext compares LLM input and output costs, context windows, and providers.
- Metatext pairs benchmark scores with cost per million tokens.
- Metatext ranks decision models by accuracy, calibration, latency, and cost.
- Metatext searches AI and NLP datasets by task, language, and benchmark status.
- Metatext lists MCP servers by transport, runtime, and author.
- Metatext lists agent skills by author and keyword.
- Metatext filters AI jobs by role, stack, location, and remote work.
How Metatext works
Metatext organizes discovery around dedicated areas for LLM pricing, benchmarks, models, datasets, MCP servers, agent skills, and AI jobs. Users can compare models by cost and context, consult benchmark leaderboards, or explore model selections by use case. Listed use cases include coding, low cost, long context, and open weights.
Metatext supports more focused searches through resource-specific browsing options. Dataset discovery uses task, language, and benchmark status; MCP server browsing uses transport, runtime, and author. Agent skills can be browsed by author and keyword, while job searches can narrow results by role, technology stack, location, and remote availability.
Who Metatext is for
Metatext is built for AI researchers, engineers, and builders making model and tooling decisions. Metatext also includes a focused AI and machine learning job board for people seeking relevant roles. The directory supports questions about which model to ship, which dataset to use, and which MCP server or agent skill to explore.
- Researchers can find datasets for training and evaluation.
- Engineers can compare model pricing against benchmark performance.
- Builders can discover MCP servers and agent skills for Claude or Cursor.
- AI job seekers can filter roles by stack, location, and remote work.
Metatext pricing and access
Metatext is free to use, and its underlying data is open. Metatext provides directory access to model pricing comparisons, benchmark results, datasets, agent resources, and job listings. Model prices displayed in comparison views describe the listed models, while Metatext itself is presented as a free directory.
Metatext strengths and data freshness
Metatext combines several types of AI resources in one searchable directory and connects benchmark performance directly with token pricing. Metatext continuously refreshes pricing, benchmark results, new models, MCP servers, agent skills, and job listings. Open underlying data and free access are additional stated characteristics, alongside resource-specific search and browsing options.
Metatext FAQ
What is Metatext?
Metatext is a free, open directory of AI models, datasets, tools, and jobs. Metatext combines LLM pricing comparisons, benchmark leaderboards, MCP server listings, and agent skills for AI researchers, engineers, and builders.
Is Metatext free to use?
Metatext is free to use, and its underlying data is open. Metatext includes model pricing comparisons, benchmarks, dataset listings, MCP servers, agent skills, and AI jobs in the directory.
How does Metatext compare LLMs?
Metatext compares LLMs by input and output cost, context window, provider, and benchmark performance. Metatext pairs benchmark scores with cost per million tokens and includes leaderboards from LMArena, LiveBench, SWE-bench, and Aider.
How current is Metatext's pricing and benchmark data?
Metatext states that pricing and benchmark results refresh continuously. Metatext also continuously updates new models, MCP servers, agent skills, and job listings, while describing the directory's underlying data as open.
Can Metatext help find MCP servers and agent skills?
Metatext lists MCP servers and packaged agent skills, including resources for Claude or Cursor. Metatext supports MCP server browsing by transport, runtime, and author, and agent skill browsing by author and keyword.
Who is Metatext for?
Metatext is built for AI researchers, engineers, and builders choosing models, training or evaluation datasets, and agent tools. Metatext also offers a focused AI and machine learning job board with role, stack, location, and remote filters.
From the community
No reviews yet
Be the first to review this tool. What you write helps the next person choose.
