Search

Smart AI decisions start with LMRank.

Compare every major model side by side. Pick an AI that fits the work you actually have.

A person exploring AI models on a phone and laptop at a warm wooden desk

Browse by category

Overall Top-ranked models across our benchmark mix. Agentic Coding Model Top picks for agentic, multi-step coding workflows. Coding Top picks for code generation, debugging, and refactoring. Long Context Models with the largest usable context windows. Open-Weight Open-weight models you can self-host or inspect. Reasoning Models tuned for step-by-step reasoning and math.

Why LMRank

An independent, transparent leaderboard built so you stop guessing which model to ship next.

500+ models tracked across every major lab
12 public benchmarks ingested on a rolling basis
Daily rankings refreshed as new models and scores land

Rankings built on the public data sources the field already trusts

  • OpenRouter Models
  • SWE-bench Benchmark
  • Artificial Analysis Benchmark
  • Exa Search

Latest articles

All articles
2026-06-24 The Rise of Model Routers: OpenRouter Fusion vs Sakana AI's Fugu Ultra 2026-06-16 Kimi K2.7 Code: Moonshot's 1T-Parameter Coding Model Hides Behind Its Own Benchmarks 2026-06-15 AI Model Weekly Roundup — June 9–15, 2026

Top 3, three ways

Switch the facet to see the leaderboard sliced by rank, by release date, or by raw input-token cost.

Stop cross-referencing a dozen provider pages. Pick a model in 60 seconds.

Compare models

Get the weekly AI digest

One email each Tuesday. New model releases, the rankings that actually moved, and a notable showcase from the week.

  • New models, ranked. No fluff.
  • The benchmarks we trust and why we trust them.
  • One real-world showcase pulled from LMBench.

Recent additions

View all models
Latest model additions and updates.
# Model Lab Added
01 Fugu Ultra Sakana 2026-06
02 North Mini Code Cohere 2026-06
03 GLM 5.2 Z Ai 2026-06
04 Kimi K2.7 Code Moonshotai 2026-06
05 Claude Fable 5 Anthropic 2026-06

About the LMRank LLM leaderboard

The LMRank LLM leaderboard tracks the major large language models so you can compare rank, pricing, context windows, release cadence, and capability fit for each task in one place. Every score is sourced from a public benchmark so you can audit the inputs and the methodology behind every ranking change.

Rankings move every week, and a single new release can reshuffle the top tools overnight. The LMRank LLM leaderboard exists so you do not have to cross-reference a dozen provider pages to decide which model is strongest, cheapest, or best aligned with the work you actually have. Browse by use case, compare tools side by side, or open any model for capability notes, context window details, and transparent per-token pricing across coding, reasoning, multimodal, and more.

LMRank pulls scores from OpenRouter, SWE-bench, Artificial Analysis, and Exa, and refreshes rankings as new models and benchmarks land. The leaderboard is independent: no pay-to-rank, no model gets a boost because its lab paid for placement. If you spot a score that looks wrong or a model that is missing, the methodology page explains how to flag it for review.