JevK5 — local AI model by alibiserikbay
Written by Jakub Rusinowski · Last updated
A 4B decision model on Qwen3.5-4B with a TypeSafe-style server and GGUF builds for llama.cpp on NVIDIA, AMD, Intel, Apple GPUs or CPU.
Variants
The smallest JevK5 variant needs about 4 GB of VRAM at Q4_K_M — quantized weights plus framework overhead, before any KV cache.
| Model | VRAM at Q4 | VRAM | Context | Run it |
|---|---|---|---|---|
| JevK5 4B → 4B | ~3.6 GB | 16K (prompt limit) | jevk5-serve :8090 |
Memory is quantized weights plus overhead at Q4_K_M, from the same engine as the GPU & VRAM checker.
How to run JevK5 locally
Install Ollama, then pull the tag.
Served by jevk5-serve (default port 8090). This model does not run in Ollama.
Pick a size above for its own VRAM figure, speed estimate and install command.
Licence
Commercial use permitted. No usage restrictions beyond attribution.
Applies to: JevK5 4BRecommended GPU
The cheapest catalogued GPU that runs JevK5 locally (min 4 GB VRAM) is the Intel Arc B570 (10 GB).
JevK5 — frequently asked questions
How much VRAM does JevK5 need?
JevK5 needs about 4 GB VRAM at Q4_K_M quantization for its smallest variant. Variants: JevK5 4B (4 GB, Q4_K_M). On Apple Silicon, unified memory counts toward this requirement.
Can I run JevK5 on an RTX 4090 (24 GB)?
Yes — JevK5 runs on an RTX 4090 (24 GB) and other 24 GB cards such as the RTX 3090. Smaller variants also fit comfortably on 8–16 GB GPUs at Q4_K_M.
What quantization should I use for JevK5?
Q4_K_M is the best balance of quality and VRAM for JevK5 in most cases. Choose Q8_0 for near-lossless quality if you have spare VRAM, or smaller quants (Q3/Q2) only when memory is tight.
How do I run JevK5 locally?
JevK5 is a decision model: it is called over an HTTP endpoint (/v1/systemone), not chatted with. Where a variant is available in Ollama 0.35 or newer, pull it with `ollama pull` and send requests to the local server; the others ship their own server. Each variant page shows the exact commands for that model.