README.md

metadata

title: llm-cal
emoji: 🧮
colorFrom: indigo
colorTo: blue
sdk: gradio
sdk_version: 6.13.0
app_file: app.py
pinned: false
license: apache-2.0
short_description: LLM inference sizing — honest, architecture-aware

llm-cal — LLM inference hardware calculator

Web UI for llm-cal. Pick a model, pick a GPU, get a hardware plan.

Architecture-aware (MLA, NSA, CSA+HCA, MoE, sliding window). Engine-aware (vLLM, SGLang). Honest-labeled — every number carries a provenance tag ([verified] / [inferred] / [estimated] / [cited] / [unverified] / [unknown]).

The story this Space exists to tell

gpu_poor reports DeepSeek-V4-Flash as 284 GB by assuming pure FP8. The real safetensors weight is 160 GB — it ships an FP4+FP8 mixed pack. llm-cal reads the actual on-disk dtype (per-tensor metadata + MX block-scaled scale tensors) and gets 160.01 GB at 0.2% error.

That's the whole pitch.

Local

pip install llm-cal gradio
python app.py

Links

• GitHub repo
• Full docs
• Methodology — every formula's primary source
• Pre-rendered model pages — popular model × GPU combos