// Personal Rankings
My AI Leaderboard
Personal tier list based on my own testing. Cloud models, local models, image models — ranked by real-world usefulness, not benchmark scores.
Top Recommended AIs
my current picks
#1
cloud
GLM-5.1
Best all-around right now. The agent sandbox on their website is a must-try — it can browse, code, and execute tasks autonomously in a live environment.
#2
cloud
DeepSeek
My go-to Google search replacement. Ask it anything you’d normally search — it reasons through answers instead of returning a list of links.
#3
cloud
Opencode
Go coding plan. If you’re serious about AI-assisted coding this is the workflow to adopt.
Cloud Coding
frontier models
#1
cloud
ChatGPT 5.5 Codex Harness
My #1 coding-agent result.
#2
cloud
Claude Opus 4.7
Strong long-task coding and repo reasoning.
#3
cloud
GLM-5.1
Strong modern coding and agent model.
Local Coding
on-device models
⚠ Hardware limited — cannot properly test larger local models right now. Rankings reflect what runs well on my current setup. Will retest with 16GB VRAM.
#1
local
Qwen3.5-9B Q8
My current best local coding model. Fast, accurate, handles multi-step tasks well.
#2
local
Gemma 4 E4B
Small reasoning and coding model worth testing locally.
#3
local
Qwopus Dual Reasoning
Finetuned coding model. Solid on logic-heavy problems.
Image Models
cloud + local
#1
cloud
GPT Image 2
My current pick for best image model overall. Strong prompt adherence, clean outputs.
#2
cloud
Nano Banana Pro
Second-place image model in my testing.
#3
local
Z-Image Turbo
Fast and useful. Good for quick iterations.
#4
local
FLUX.2 [klein] 9B
Good model but ~30% body deformation rate in my tests. Dropped from regular use until it improves.
// Last updated: May 2026