I’ve Lost 60% of My Leads in 7 Months — Here’s What I’m Doing Wrong
I’ve Lost 60% of My Leads in 7 Months — Here’s What…
Quantization for 8GB VRAM: Q4_K_M vs IQ4_XS vs UD-IQ4_NL
Quantization is the process of reducing a model’s precision from 16-bit floating…
9 Models Tested, 3 Rejected: What I Learned From 32 Benchmarks
Not every model that claims to be “good for coding” actually is.…
My 3-Model Local AI Stack: Fast Worker + Coder + Senior Planner
I used to run one model for everything. ChatGPT for code generation,…
Gemma 4 vs Qwen vs Ornith: Which Local LLM Codes Best on 8GB
I tested three model families on the same hardware, same quantization, and…
turbo2 vs turbo3 vs turbo4: Which KV Cache Quantization Fits Your VRAM
When I first tried running a 35B model at 128k context on…
Speculative Decoding Explained: How MTP Turns 18 t/s Into 36 t/s
I was stuck at 18 tokens per second on my local 35B…
Asymmetric KV Cache: Why Compressing V More Than K Unlocks 128k
Every local LLM guide tells you to pick a quantization level for…
Ornith 35B on 8GB VRAM: 128k Context at 30 t/s
I tested every model that claims to be good for local coding.…
I Fit a 35B Model Into 8GB VRAM. Heres the Full Setup.
I spent three weeks trying to run large language models locally on…
