Find the right LLM for the job.
Scores, pricing, and context side by side, so you can pick the right model without opening a dozen provider pages.
Top models right now
Full leaderboardBrowse by category
All categories
Overall
Top-ranked models across our benchmark mix.
Coding
Top picks for code generation, debugging, and refactoring.
Agentic Coding Model
Top picks for agentic, multi-step coding workflows.
Open-Weight
Open-weight models you can self-host or inspect.
Cheap Model
Strong models with the lowest per-token pricing.
Multimodal
Models that handle image, audio, and text inputs.
Latest articles
View all
Recent additions
View all models| # | Model | Lab | Added |
|---|---|---|---|
| 01 | T Thinking Machines: Inkling | Thinkingmachines | 2026-07 |
| 02 | Muse Spark 1.1 | Meta | 2026-07 |
| 03 | Kimi K3 | Moonshot | 2026-07 |
| 04 | KAT-Coder-Air V2.5 | Kwaipilot | 2026-07 |
| 05 | KAT-Coder-Pro V2.5 | Kwaipilot | 2026-07 |
About the LMRank LLM leaderboard
LMRank tracks 120 large language models from 34 labs. Rankings pull from OpenRouter, SWE-bench, Artificial Analysis, and Exa, refreshed daily. Every score traces to a public source you can audit.
The LLM leaderboard moves as new models ship and new results land, so a single release can reshuffle the top tools overnight. No lab pays for placement, and no model gets a boost because its vendor sponsored the site.
Start with the full leaderboard, browse categories like coding and reasoning, or compare two models side by side. If a score looks wrong, the methodology page explains the process and how to flag a correction.