LLMs
10 Lakh Token Ke Baad Bhi AI Bhool Jaata Hai — 3 Wajah
aicreatorhub.netGuide
1 Lakh Stars Wali Repo: AI Andar Se Seekhiye, 3 Wajah
ChatGPT jaisa model khud banate hue seekhiye — step by step, PyTorch mein. 104,158 stars. Ye course nahi, code hai — aur yahi iski taakat aur iski shart dono hai.
AI Jhooth Kyun Bolta Hai? 3 Wajah
AI Benchmark Ke 3 Jhooth: 95% Score Ka Matlab
AI galat jawab poore confidence ke saath deta hai — ye bug nahi, design hai. Teen wajah, aur wo ek sawaal jo 90% galat jawab pehli hi baar mein pakad leta hai.
Har naya model kisi na kisi benchmark mein number one hota hai. Teen wajah jinse ye aankde dhoka dete hain, aur wo ek test jo aapke liye sach mein maayne rakhta hai.
Lambi chat mein AI shuruaat bhool jaata hai — iski wajah context window hai. 1M token tak ke model aa chuke hain, phir bhi ye samasya rehti hai. Kyun, aur kya karein.
10 Lakh Token Ke Baad Bhi AI Bhool Jaata Hai — 3 Wajah
Log sochte hain AI 'yaad rakhta hai'. Nahi rakhta. Har message par poori baat-cheet phir se uske saamne rakhi jaati hai, aur jo us limit mein nahi aata wo hat jaata hai — chup-chaap, bina bataye.
Model shabd nahi, token padhta hai. Ek angrezi shabd aam taur par ek token hota hai; Hindi aur Hinglish mein ek hi shabd kai tokens mein tootta hai. Isliye same baat Hindi mein zyada jagah leti hai — aur limit pehle aa jaati hai.
Ye sabse zaroori baat hai. Jab chat window se badi ho jaati hai, to shuruaat ka hissa hata diya jaata hai — aur AI aapko batata nahi. Isliye lagta hai ki wo 'bhool gaya', jabki asal mein wo hissa use dikhaya hi nahi gaya.
| Kya hota hai | Aapko kaisa lagta hai |
|---|---|
| Shuruaat ka hissa hat gaya | "Maine to shuru mein bataya tha" |
| Instruction hat gaya | "Ye phir se angrezi mein likhne laga" |
| File ka aadha hissa hat gaya | "Aadhe document ka jawab galat hai" |
Ab 1M token tak ke context wale model maujood hain, jo lagbhag ek poori kitaab hai. Lekin bada window ka matlab behtar dhyan nahi hota: beech mein rakhi hui baat par model ka dhyan sabse kam jaata hai. Zaroori instruction shuru mein aur ant mein rakhna abhi bhi kaam karta hai.
Nayi chat shuru kijiye
Naya kaam, nayi chat. Purani chat ghaseetne se sirf jagah bharti hai.
Zaroori instruction dobara likhiye
Lambi chat mein use phir se upar rakh dijiye.
Beech ka kachra hataiye
Jo aazmaish fail ho gayi, use lekar aage mat chaliye.
Lambe document tukdon mein dijiye
Poori kitaab ek saath se behtar hai chapter dar chapter.
Aam anumaan: angrezi mein 4 characters ka ek token, Hindi mein iska aadha ya usse bhi kam. Zyadatar providers apna token counter dete hain.
Us chat ka context to jaata hai. Lekin kya provider aapka data rakhta hai, ye alag baat hai — uske apne settings dekhiye.
Lambe document ke saath kaam karte waqt angrezi mein poochhna tokens bachata hai. Chhoti baat-cheet mein farak nahi padta.
Save this summary as an image or share it.
AICreatorHub Team
The AICreatorHub editorial team is a group of hands-on AI practitioners, writers and developers based in India. We test AI tools and models ourselves, track official releases from OpenAI, Anthropic, Google, Meta and xAI, and translate them into simple, India-first guides in English and Hindi. Every article is written for real Indian use cases — pricing in rupees, free-tier tips and practical, tested steps — so you get accurate, up-to-date and genuinely useful AI information.