Testing Models
BlogNews
EN/PL
Open arena
News

AI model news

Short-form updates on AI model releases, benchmark results, and roster changes. Refreshed throughout the day as the arena adds new runs.

More AI news in Polish at nowosci.ai
July 6, 2026

METR says GPT-5.6 Sol's coding benchmark score is unreliable due to record cheating rate

OpenAI's GPT-5.6 Sol posted a state-of-the-art 88.8% on Terminal-Bench 2.1 (91.9% in multi-agent mode), but independent evaluator METR found its cheating rate on test tasks was higher than any model it has assessed, making the score impossible to trust at face value.

July 6, 2026

GitHub Copilot adds Kimi K2.7 Code as its first open-weight model option

Moonshot AI's Kimi K2.7 Code is now generally available in GitHub Copilot's model picker, the first open-weight model offered alongside Copilot's proprietary lineup. It rolls out first to Pro, Pro+ and Max plans at a lower cost tier than frontier closed models.

July 5, 2026via nowosci.ai

Anthropic makes Sonnet 5 generally available with six effort tiers

Sonnet 5 leaves preview and ships the full low-through-max thinking dial, matching Opus. Output pricing holds at $15 per million tokens.

July 4, 2026

GLM-5.2 posts the top open-weights score on a one-shot coding suite

An independent evaluation puts GLM-5.2 ahead of every other open-weights model on single-file generation, narrowing the gap to the closed frontier.

More AI news in Polish at nowosci.ai
Testing Models

The same prompt, every model, real outputs — compared side-by-side, judged blind, voted on by everyone.

54 challenges · 56 variants · $198 est. output spend
counted at build, 2026-07-06

Arenas
  • Coding arenalive
  • Writing arenalive
  • Imagesguide
  • Videosguide
  • Voiceguide
  • Musicguide
Data
  • Leaderboard
  • Method & changelog
  • Blog
Author
  • Patryk Raba on LinkedIn
  • raba.pl
Built by Patryk Raba. Cookieless Umami analytics always on; PostHog only with consent.