FrontierCode Leaderboard
FrontierCode Leaderboard is Cognition's FrontierCode 2.0 data has a hidden model, "Obsidian 3": 46.8%, above #1 Opus 5.5 (45.5%). It runs in the claude-code harness, which only Claude mode. It is ranked #154 on the AI Radar, in Frontier models, first seen 31 days ago and shared in 16 posts (719.8k views).
What FrontierCode Leaderboard says about itself
Live rankings of frontier AI models on FrontierCode, Cognition’s benchmark for production-quality code. Compare scores, pass rates, cost, and speed.
FrontierCode Leaderboard | Cognition Menu Close 01 FrontierCode Leaderboard By The Cognition Team Read the methodology Explore a task FrontierCode 2.0 Score vs Cost Scale Log Linear Score is a weighted aggregate of the rubric items. Solutions that don’t pass blocking criteria receive 0. Cost is the mean USD spend per rollout at list API pricing, including public discounts. Each line connects a model’s runs across reasoning efforts. The most efficient models sit top-left. The x axis uses a log scale. Models ▲ 02 Methodology FrontierCode is the first benchmark to measure mergeability: would the maintainer actually merge this PR? Our criteria…
What people said about FrontierCode Leaderboard on X
We've evaluated Gemini 4 Carbon and Gemini 4 Argon on FrontierCode 2.0, our coding benchmark that measure's a model's ability to go beyond the spec. Gemini 4 Carbon scores near SOTA with a score of 40% and a cost of $4.04 per task, behind Claude Opus 5.5, GPT-6 Astra, and Claude Sonnet 5.5. Gemini 4 Argon scores…
— @cognition, 7 days ago · 1.2k likes · see the post
🚨 Obsidian 3 is hidden model in FrontierCode 2.0 This is the highest scoring model : what can it be < a Fable 5.5 or something else> Harness: it runs on claude-code, which only the six Claude models use Price level: it costs about 3.6× what Opus 5.5 costs at the same effort.
— @chetaslua, 9 days ago · 652 likes · see the post
Google might actually be back
— @scaling01, 7 days ago · 577 likes · see the post
Alternatives to FrontierCode Leaderboard
- Thinking Machines Lab — Mantic have been using Tinker since it launched. This guest post is a technical deep dive on what they have built so…
- openai.com —
- Tencent Hy — Tencent Hy
- clocktowerbench.com — You can see the full benchmark at the website, and I’ve also made it so that you and your friends can play again the AIs
- Claude — Basecamp for Claude is live!
- Mercor — Evaluate frontier AI models with Mercor's APEX AI Benchmarks & Leaderboards. Measure AI productivity across software…
FrontierCode Leaderboard in numbers
- Rank on the AI Radar: #154 of 10389
- Shared in 16 posts by 16 accounts: @cognition, @chetaslua, @scaling01, @ChrisGPT
- 719.8k views on those posts
- First seen 31 days ago, last shared 5 days ago
- Pricing seen by AGK Intelligence: not stated
- Market: Frontier models
FAQ
What is FrontierCode Leaderboard?
Cognition's FrontierCode 2.0 data has a hidden model, "Obsidian 3": 46.8%, above #1 Opus 5.5 (45.5%). It runs in the claude-code harness, which only Claude mode It was first shared on X 31 days ago and is ranked #154 on the AI Radar.
Is FrontierCode Leaderboard free?
Pricing is not stated on the page we read.
Who shared FrontierCode Leaderboard?
16 accounts on X, including @cognition, @chetaslua, @scaling01, in 16 posts totalling 719.8k views.
Every post is read and classified by AGK Intelligence: what it is, which market it belongs to, and whether the link is a real tool. 65.7k posts from 5k X accounts over the last 14 days, 10.4k tools, 19 markets. New posts are collected several times an hour and the full ranking is recomputed every hour — last update 2026-11-06 13:30 UTC. Full method.