AI model intelligence, free and open
Explore 40+ models from 14 providers — compare pricing, count tokens, estimate costs, check rankings, and follow AI news. No account, no ads, no paywall.
- 100
- Models
- 14
- Providers
- 7
- Languages
- 34,650+
- Comparison pages
Top Models Leaderboard
Transparent scores from public benchmarks, capability, price, and context.
- 1 Claude 3.5 SonnetAnthropic78.4Score90%Data ConfidenceNeeds verificationLast verified2Sources
- 2 Gemini 1.5 ProGoogle76.9Score90%Data ConfidenceNeeds verificationLast verified2Sources
- 3 Mistral Large 2Mistral73.6Score90%Data ConfidenceNeeds verificationLast verified2Sources
- 4 GPT-4oOpenAI73.1Score90%Data ConfidenceNeeds verificationLast verified2Sources
- 5 Llama 3.3 70BMeta63.0Score90%Data ConfidenceNeeds verificationLast verified1Sources
- 6 Qwen2.5 72BAlibaba62.6Score90%Data ConfidenceNeeds verificationLast verified1Sources
- 7 Gemini 1.5 FlashGoogle60.9Score90%Data ConfidenceNeeds verificationLast verified1Sources
- 8 Claude 3 OpusAnthropic60.3Score90%Data ConfidenceNeeds verificationLast verified1Sources
Updated dailyBased on public benchmarks (MMLU, GPQA, HumanEval, AIME) and official pricing. See the full leaderboard for details.
Ranking Methodology
- Score calculation
- Benchmark 50% · Capability 20% · Price 20% · Context 10%
- Benchmark sources
- HumanEval · MMLU · GPQA · AIME · LMSYS Arena
- Pricing sources
- OpenAI · Anthropic · Google official pricing
- Last update
- Daily snapshot (UTC 02:00)
Explore the platform
AI Models
Browse 49 models across 14 providers with context windows, capabilities, and per-1M-token pricing. Filter by provider, search, and sort.
→
Compare Models
Side-by-side comparison of any two models: capabilities matrix, pricing, benchmark scores, and ranking history.
→
Leaderboard
Transparent scoring across benchmark, capability, price, and context — updated daily with trend history.
→
Benchmarks
Transparent scoring across benchmark, capability, price, and context — updated daily with trend history.
→
Use Cases
Scenario-based model picks for coding, reasoning, math, vision, writing, chatbots, long context, and budget APIs.
→
Recommendations
Transparent scoring across benchmark, capability, price, and context — updated daily with trend history.
→
AI News
Daily aggregated AI industry news from major labs, with RSS feed and multi-language pages.
→
How it works
- 1
Pick a model or two from the database
- 2
Compare specs, prices, and benchmarks side by side
- 3
Estimate your token budget and costs
- 4
Follow the leaderboard and daily AI news
Transparent by design
Every price carries its source and verification status. Data comes from official pricing pages where possible, and unknown values are clearly marked — no ads, no affiliate links, no paywall, ever.
Frequently Asked Questions
What is the best AI model?
Based on our transparent scoring, Claude 3.5 Sonnet currently leads with an overall score of 78.4 out of 100. Scores combine public benchmarks, capability, price efficiency, and context window, and are updated daily.
How are AI models ranked?
Models are scored on a transparent 0-100 formula: public benchmarks (50%), capabilities (20%), price efficiency (20%), and context window (10%). Rankings are snapshotted daily; the full methodology is documented on the ranking page.
How often are prices updated?
Prices are sourced from official provider pricing and refreshed on a regular schedule, with price history tracked per model. The last update date is shown on each model page.
Where does benchmark data come from?
Benchmark results come from public sources (e.g. MMLU, GPQA, HumanEval, AIME) with dataset, version, and test date attributed on each entry. Note that some entries are community-submitted and unverified, as marked.
Ready to explore?
The platform is free and open. Start with the model database or jump straight into the tools.