📰 Dev.to · Dev Yadav
10 articles · Updated every 3 hours · View all reads
All
Articles 90,761Blog Posts 108,826Tech Tutorials 22,671Research Papers 19,220News 14,833
⚡ AI Lessons
Dev.to · Dev Yadav
2mo ago
The Demo Was One User. Then Batch Size Became Real.
The demo worked because the test was one user, one prompt, one response. Then real usage showed up,...
Dev.to · Dev Yadav
2mo ago
4-bit Quantization Does Not Make VRAM Problems Go Away
A lot of people hear 4-bit quantization and mentally convert that into this model should run anywhere...
Dev.to · Dev Yadav
2mo ago
KV Cache Is Why Your Model Fit Until It Did Not
The model loaded. The first prompt worked. Then longer prompts or multiple users showed up, and...
Dev.to · Dev Yadav
2mo ago
7B Parameters Does Not Mean 8GB VRAM Is Enough
A lot of people see 7B and assume 8GB VRAM should be enough. Then they load the model, increase...
Dev.to · Dev Yadav
2mo ago
The Model Was Cheap. The Retries Became the Bill.
The hourly price did not look scary. What hurt was running the same job again, reloading the same...
Dev.to · Dev Yadav
2mo ago
The Tutorial Used Tiny Prompts. Your Real Prompts Did Not.
The tutorial looked smooth because the prompt was tiny. Then you used the real prompt your app...
Dev.to · Dev Yadav
2mo ago
The Demo Worked on a 7B Model. Production Traffic Changed the Math.
The demo looked fine on a small model with one user. Then real traffic showed up, latency got ugly,...
Dev.to · Dev Yadav
2mo ago
The Cheapest GPU Looked Smart. Then the Job Took All Night.
The hourly price looked great, so the cheapest GPU felt like the responsible choice. Then the run...
Dev.to · Dev Yadav
2mo ago
Your Model Loaded Fine. Then Context Length Broke the GPU Plan.
The model loaded. The notebook worked. Then you increased context length, batch size, or both, and...
Dev.to · Dev Yadav
3mo ago
Best GPU Rental for AI Training in India
Training LLMs? Fine-tuning models? Here's which GPU you actually need (and which ones are...
DeepCamp AI