Sabitlenmiş Tweet
Shoriful Dev 🔲
1.9K posts
Shoriful Dev 🔲
@shoriful_dev
Full-Stack Engineer | FREE AI Tools + Pro Tech Tips | Building in Public | Ship faster with AI in 2026 📈
Newsletter Katılım Ağustos 2024
698 Takip Edilen1K Takipçiler
AI is no longer just a "buzzword," it's the new electricity. ⚡️
In the next 5 years, AI won't replace humans. But humans using AI will replace those who don't. 🤖💡
Are you upskilling yourself for the AI era? Let me know below! 👇
English
Best AI models by your use cases : 🤝
- Frontend : Kimi-K3
- Backend : Fable
- Debug : GPT-5.6-SOL
- Image gen : GPT
- Translation: Gemini-3.5-Flash
- Search : Grok-4.5
- Video gen : Seedance 2.0
- Best price : Deepseek V4 Pro
- Local AI (Large) : GLM-5.2
- Local AI (Small, 128gb) : HY-3, DSV4 Flash
Use models accordingly
English
@theo Weights ≠ harness. Sol topping Artificial Analysis's coding index is on Codex's own harness — swap it into Claude Code's orchestration via proxy and you're testing a different system entirely. Anyone benchmarked the same task both ways?
English
"Why do you use gpt-5.6 and kimi k3 inside of Claude Code?"
This vid goes in depth on what Claude Code gets right, what Codex gets wrong, and how to copy my setup :)
Blessed by Tibo btw x.com/thsottiaux/sta…
Tibo@thsottiaux
If you aren't yet bold enough to install the Codex app, you can stay in the presence of your orange crab and point it at GPT 5.6 Sol. Takes 5 minutes. Kudos to Theo for explaining one of the ways to get this done. Step 1: Install CLIProxyAPI Step 2: Connect Step 3: Define following alias and enjoy claudex ``` alias claudex='CLAUDE_CODE_SUBAGENT_MODEL=gpt-5.6-sol \ CLAUDE_CODE_ALWAYS_ENABLE_EFFORT=1 \ CLAUDE_CODE_MAX_TOOL_USE_CONCURRENCY=3 \ ENABLE_TOOL_SEARCH=false \ claude --model gpt-5.6-sol' ``` If this gets blocked, I owe you a reset.
English
Shoriful Dev 🔲 retweetledi
12 AI tools you need to try in 2025:
1. ChatGPT.com (solves anything)
2. RecCloud.com (speech to text)
3. MidJourney (generates art)
4. Replit (generates code)
5. Synthesia (generates video)
6. Soundraw(generates music)
7. Fliki (generates TikToks)
8. Starry(generates avatars)
9. SlidesAI (generates PPTs)
10. PicWish (edit pictures)
11. Pictory (edit videos)
12. Relume (web designs)
English
Shoriful Dev 🔲 retweetledi
@steipete @clawsweeper Matches what OpenAI's own data shows — xhigh vs high on Terra barely moves accuracy but tanks latency, so you're paying for thinking tokens that don't convert to better judgment on scoped tasks like review. Right call stopping at high.
English
5.6 Terra high is underrated. Switched @clawsweeper (GitHub review bot) to it and it's ~40% faster overall with negligible quality loss. Better than 5.5 on all counts. Massively cheaper.
(Tried xhigh but that negates perf wins, didn't make a noticable difference in review evals)
English
Makes sense why they're temping this — K3 was already the outlier on Go's usage table (~340 req/wk vs 2k+ for GLM-5.2), since it doesn't get the bulk-discount multiplier others do. 2x for a week ≈ their normal monthly allotment. Curious if this becomes permanent or just a stress test.
English
Just did some research yesterday, someone commented, "Kimi is good for Indonesian language content" turns out it's above Fable , just found out, even though the price is below it
English
@OpenRouter @OpenAIDevs This exact walk-vs-drive puzzle format has been in eval corpora since GPT-4 — testing it just measures memorization, not reasoning. Try it with scrambled units (50m → 50mi) and watch Sol/Terra/Luna all break the same way.
English
@steipete Open source work streamed live during OpenAI Build Week is a fun crossover. What's the repo?
English
@claudeai Learning Commons connector is the real story here — Claude checking lesson plans against actual state standards beats another chatbot that just writes worksheets. Anyone testing it against ChatGPT for Teachers yet?
English
We're introducing Claude for Teachers: free access to premium Claude capabilities for verified K-12 educators in the US, with a library of teaching skills and a direct connection to evidence-based curricula, mapped to academic standards in all 50 states.
claude.com/solutions/teac…
English