Mainbrella Build: a Lovable-style app builder running open models on Cloudflare Workers AI
CodeRead the post
mainbrella.com · web source · link re-checked 2026-10-16
Kimi K2.5
By Moonshot. 2 results on 2 boards, copied as published. Different boards, different metrics: don't add them up.
License (as listed)Open weights
Hosting regionNot confirmed
Boards2
Code
- 70.8%SWE-bench Verified (Bash Only)
Kimi K2.5 (high) · % Resolved
mini-SWE-agent 2.0.0 (same environment for all) · as of 2026-02-17 Stale ↗
Agentic / terminal
No result on the boards we captured
Websites
- 1114Design Arena (Fullstack)
kimi-k2.5 · Elo · board #34
Design Arena · as of 2026-10-23 ↗
Design
No result on the boards we captured
Android
No result on the boards we captured
iOS
No mature public leaderboard
Available in
No plan row in our pricing data names this model version.
Use cases with this model
No sourced use case names this exact model version.
Same family: Kimi 15
MotionPromptGallery: "178 AI motion graphics … made with Opus 5.5, GPT-6 Astra, Kimi K3, Fable 5", Next.js on Vercel
WebsitesDesignRead the post
x.com · link re-checked 2026-10-16
Cline + Kimi K3 evaluated in production and on Terminal-Bench 2.1 (pass rate 77.5% to 88.8%)
CodeRead the post
dev.to · web source · link re-checked 2026-10-16
Test: "Qwen 3.8 Max (max): 25.7 - DeepSeek V4.1 Flash (max): 21.7 - Kimi K3 (default): 23 - GLM-5.3 (max): 19 - Mistral Large 4 (high, its max): 15"
CodeRead the post
x.com · link re-checked 2026-10-16
"benchmarked worse + similar price as Kimi K3 and GLM 5.3 in the Pi harness", run on real PRs
CodeRead the post
x.com · link re-checked 2026-10-16
Same 3D prompt one-shot across 8 models incl. Mistral Large 4
DesignRead the post
x.com · link re-checked 2026-10-16