VOOZH about

URL: https://apify.com/artful_sleep/lm-bench

⇱ LM Bench Β· Apify


πŸ‘ LM Bench avatar

LM Bench

Under maintenance

Pricing

Pay per usage

Go to Apify Store

LM Bench

Under maintenance

A daily-updated intelligence feed that merges LLM quality (LMArena ELO) and LLM pricing (OpenRouter) into one clean, normalized dataset β€” so you can always choose the best model for your task with best price daily.

Pricing

Pay per usage

Rating

0.0

(0)

Developer

πŸ‘ Yue Yang

Yue Yang

Maintained by Community

Actor stats

0

Bookmarked

2

Total users

0

Monthly active users

7 months ago

Last modified

Share

πŸš€ Daily LLM Intelligence

A daily-updated intelligence feed that merges LLM quality (LMArena ELO) and LLM pricing (OpenRouter) into one clean, normalized dataset β€” so you can always choose the best model for your task.


❓ What’s the problem?

Every AI app says:

β€œWe use the best model for your task.”

But in reality, most apps are NOT using the best model:

  • Many pay 10Γ— more for a model that’s not actually better.
  • Many use outdated choices because pricing and performance change every week.
  • Everyone thinks they know which LLM is best β€” but no one actually tracks it daily.

So companies waste money, use wrong models, and miss cheaper or better options.


🧩 What do we have today?

We already have:

  • Quality data from LMArena (ELO scores)
    β†’ but it’s not connected to pricing.

  • Pricing data from OpenRouter
    β†’ but the model names don’t match.

  • Many model variants & providers
    β†’ but no unified naming.

So although the data exists, it is fragmented and unusable for real decisions.


βœ… What did I build?

I built Daily LLM Intelligence β€” an Apify Actor that:

  • βœ” Tracks LLM quality daily (from LMArena)
  • βœ” Tracks LLM pricing daily (from OpenRouter)
  • βœ” Normalizes model names (fixes the Qwen / Qwen2.5 / 72B chaos)
  • βœ” Merges everything into one clean dataset

Each row = one canonical model with:

  • Quality score (ELO)
  • Price (input / output)
  • Provider
  • License
  • Rank by quality
  • Rank by price
  • Value ratio

🧠 What can users do with it?

With this dataset, you can:

  • Automatically pick the best-value LLM
  • Avoid overspending
  • Route traffic to cheapest high-quality models
  • Build dashboards or alerts
  • Build an auto model router
  • Pair it with a recommendation agent

It turns LLM selection into a data-driven decision, not guesswork.


βš™οΈ How does it work?

  1. Fetch LMArena leaderboard
  2. Fetch OpenRouter pricing
  3. Normalize model names
  4. Merge into one clean table
  5. Compute basic ranks
  6. Output daily dataset

Fully automated.


🧱 What’s inside the Actor?

  • LMArena scraper
  • OpenRouter pricing fetcher
  • Model name normalizer
  • Merge engine
  • Rank calculator
  • Dataset writer

Minimal. Stable. Production-oriented.


⚑ 10-second Version

Most AI apps use the wrong model and waste money.
Quality and pricing change constantly, but no one tracks them daily.

Daily LLM Intelligence merges real quality + real price + name normalization into one clean dataset β€” so you can always pick the best model today.

You might also like

OpenRouter Models Scraper - LLM Pricing and Catalog

gio21/openrouter-models-scraper

Scrape the OpenRouter LLM model catalog: pricing, context length, providers, capabilities.

Best Actor Finder

pranavpatel/best-actor-finder

Finds and tests the best actors for a specific task.

19

5.0

OpenRouter - Bring Your Own Key

superlativetech/open-router

Use the OpenRouter platform to choose the best and most cost effective model for your prompts utilizing a standardized interface (including ChatGPT, Claude, Gemini, Llama, Mistral, and more). See instructions for creating an OpenRouter account and API key.

LMArena LLM Leaderboard Scraper

jungle_synthesizer/lmarena-llm-leaderboard-scraper

Scrape the LMArena (Chatbot Arena) ELO leaderboard β€” ranks, ratings, vote counts, and confidence intervals across all arena variants (text, code, vision, document, image, video, and more). Returns one row per model per leaderboard variant.

πŸ‘ User avatar

BowTiedRaccoon

2

LLM API Pricing Monitor & Tracker

devilscrapes/llm-pricing-monitor

Scrape and compare live LLM API pricing from OpenAI, Anthropic, Google, Mistral, Groq, Together AI, and DeepSeek β€” normalized per-million-token, export to JSON or CSV. A continuously updated LLM API pricing comparison table for cost dashboards and FinOps.

OpenRouter Model Scraper

datapilot/openrouter-model-scraper

OpenRouter Models Scraper extracts AI model metadata from OpenRouter API, including pricing, context length, providers, modalities, token limits, vision/tool support, JSON support, and model architecture. Supports keyword filtering, proxy rotation, and structured dataset

OpenRouter

apify/openrouter

You can use any AI LLM model without accounts in AI providers. Use this Actor as a proxy for all requests. Use pay-per-event pricing to pay only for the real credit used.

OpenRouter LLM Model Pricing Scraper

jungle_synthesizer/openrouter-llm-model-pricing-scraper

Fetch the full OpenRouter model catalog with pricing, context length, capabilities, and supported parameters as structured data. Returns all 350+ LLMs with prompt/completion/cache prices, modalities, tokenizer details, knowledge cutoffs, and per-provider routing links.

πŸ‘ User avatar

BowTiedRaccoon

4

LLM Radar - AI Model Pricing, Benchmarks & Status Actor API

datahq/llm-radar

Real-time pricing for 110+ AI models, live LMSYS Arena ELO scores, and provider operational status from 11 providers. One API call.

DataHQ

18

Related articles

Extract contact details from Google Maps and enrich it in one run
Read more
AI web scraping and automation with Python
Read more
How to build a sentiment analysis pipeline for customer feedback
Read more