AIStupidlevel
@AIStupidlevel
Followers
41
Following
16
Media
10
Statuses
38
AI gets smart. AI gets stupid. We measure the difference. Official AI Stupid Level benchmarks. Open Source. A product of @studioplatforms
Europe
Joined September 2025
♕ AI Model Rankings - Live Performance Scores: ① GPT-4O-2024-11-20 - 71 pts ② CLAUDE-OPUS-4-1-20250805 - 69 pts ③ KIMI-LATEST - 69 pts Real-time AI intelligence monitoring Updated: Oct 16, 2025 at 01:12 PM UTC
aistupidlevel.info
Compare AI models with our comprehensive benchmarking tool. Test Claude vs GPT vs Gemini performance. Find the best AI for coding and development.
0
0
0
♕ AI Model Rankings - Live Performance Scores: ① GPT-4O-2024-11-20 - 70 pts ② CLAUDE-3-7-SONNET-20250219 - 69 pts ③ GROK-CODE-FAST-1 - 69 pts Real-time AI intelligence monitoring Updated: Oct 16, 2025 at 10:31 AM UTC
aistupidlevel.info
Compare AI models with our comprehensive benchmarking tool. Test Claude vs GPT vs Gemini performance. Find the best AI for coding and development.
0
2
2
PRO plan is LIVE Smart Router - All your keys in -> Best performing out Compare - Deeper analytics New Models - KIMI - DEEPSEEK - GLM
0
2
4
CLAUDE-SONNET-4-20250514 demonstrated noticeable degradation during the most recent evaluation. It is recommended to avoid this version.
0
1
2
Update Added 95% confidence intervals and reliability badges to all AI model scores. You can now see which models are consistent vs. unpredictable in their performance. Also improved site speed with Redis caching and database optimizations.
0
4
5
If your favorite AI model starts acting up, we catch it in real time so you can jump to one that’s working like it should.
0
1
3
Just got featured on @StirileProTV “I Like IT” to present @AIStupidlevel The first platform that tracks AI performance & drift in real time and it’s already helping thousands worldwide. Full story 👉
stirileprotv.ro
Un român a creat o platformă care măsoară performanța inteligenței artificiale în timp real. Cum funcționează
0
4
7
Is your AI acting dumb today? 🤔 You’re not imagining it. Models really drift. See which one’s “stupid” right now 👉 https://t.co/pEywIPGKyu
0
4
4
Big drop: TOOL-CALLING is live. Models now run real tools, not just talk. Intelligence Center got a brain transplant. Come break it:
0
2
1
ACTIVE DEGRADATIONS GROK-4-0709 (xAI): Performance dropped 18% (72 → 59) SMART PICKS (CODING focus) Best for Code → GPT-5-NANO (#1, 73% stability) Most Reliable → Claude 3.5 Sonnet (80% consistency) Fastest → GPT-5-NANO (1000ms avg) AVOID GROK-4-0709-EU & Latest
1
3
3
All models are stable. Now is the perfect time to start working on your project.
0
1
2
Claude is x2 more expensive than Gemini, same performance. gemini-2.5-pro-preview-03-25: score=74 | corr=99.1% | lat~6572ms | ~$0.109 claude-sonnet-4-20250514: score=74 | corr=99.1% | lat~3039ms | ~$0.198
0
2
3
@AIStupidlevel got featured in NotebookCheck!! Read the full article here
notebookcheck.net
A new open-source tool is offering real-time monitoring of multiple AI models, including OpenAI GPT-5, Claude Opus 4, and Gemini 2.5 Pro. The first of its kind, it can detect "when AI companies...
1
4
5
The new open-source Qwen3-Next Instruct and Thinking models put state-of-the-art long-context reasoning into the hands of everyone. We collaborated with #opensource frameworks from SGLang (@lmsysorg) and @vllm_project to enable communities to deploy Qwen3-Next across the
9
34
133