LLM leaderboard changes
Briefpanel reads the main public LLM leaderboards once a day, Artificial Analysis, OpenRouter rankings, Scale's SEAL leaderboards, Vellum and SWE-bench, and records what changed: a new model on the board, a score or a rank that moved. Every item below comes straight from the leaderboard's own page and shows when that page was last checked.
Each leaderboard measures something different. Read its method on its own page before you compare scores across boards.
5 sources40 changes in the last 7 days
Get an alert when the leaderboard you follow moves
Get an alert when something changes on the page you pick.
No credit card required · Set up in 10 minutes
Fri, Oct 30
40 itemsNew
MiniMax M3.1 Flash
New
Hy4 preview
New
Claude Opus 5.5
New
DeepSeek V4.1 Flash 1017
New
GLM 5.3
New
DeepSeek V4 Flash 0731
New
GPT-6 Luna
New
GLM 5.3 Flash
New
Claude 4.5 Opus | Sonar Foundation Agent
Claude 4.5 Opus powered by Sonar Foundation Agent scored 79.20% resolved. Evaluation date 2025-12-05.
New
Gemini 3 Pro Preview | live-SWE-agent
Gemini 3 Pro Preview powered by live-SWE-agent scored 77.40% resolved. Evaluation date 2025-11-20.
New
Claude Sonnet 4 | Harness AI
Claude Sonnet 4 powered by Harness AI scored 74.80% resolved. Evaluation date 2025-07-31.
New
MiniMax M2.5 high | mini-SWE-agent
MiniMax M2.5 high powered by mini-SWE-agent scored 75.80% resolved. Average cost $0.07. Evaluation date 2026-02-17. Release 2.0.0.
New
Claude 4 Sonnet | EPAM AI/Run Developer Agent
Claude 4 Sonnet powered by EPAM AI/Run Developer Agent scored 76.80% resolved. Evaluation date 2025-08-04.
New
Claude 4.6 Opus | mini-SWE-agent
Claude 4.6 Opus powered by mini-SWE-agent scored 75.60% resolved. Average cost $0.55. Evaluation date 2026-02-17. Release 2.0.0.
New
Claude 4.5 Opus medium | live-SWE-agent
Claude 4.5 Opus medium powered by live-SWE-agent scored 79.20% resolved. Evaluation date 2025-12-15.
New
Gemini 3 Flash high | mini-SWE-agent
Gemini 3 Flash high powered by mini-SWE-agent scored 75.80% resolved. Average cost $0.36. Evaluation date 2026-02-17. Release 2.0.0.
New$2
GPT-6 Sol
GPT-6 Sol is listed in the Model Comparison table with 1,050,000 context size and $2 input / $10 output cost.
New$0.04
Nova Micro (Cheapest)
Nova Micro is ranked first in the Cheapest Models (per 1M tokens) chart with $0.04 input and $0.14 output per 1M tokens.
New
Best Overall (Humanity's Last Exam)
Benchmark chart ranking models by score on Humanity's Last Exam, with Claude Opus 5.5 leading at 67.7%.
New
Best in Agentic Coding (SWE Bench)
Benchmark chart ranking models by SWE Bench score, with GPT-5.6 Sol leading at 96.2%.
How this radar works
Straight from the source
Every item is read from the source page and links back to it. Summaries are automatic; when in doubt, the source is what counts.
Automatic checks
Every page is checked automatically and often. The time of the last check shows on each source.
What changed, in one line
When an item that was already published changes, it shows as Updated, with the difference summarized.
The source page prevails. This summary may be out of date or incomplete: always check the source before acting.
Track the pages that matter to you
BriefPanel reads the page you pick, as often as you choose, and shows the before and after of every field that changed.
No credit card required · Set up in 10 minutes