The hardware roster · one machine
Rig B — the gaming laptop
MSI·whole machine·14 figures, from 2 exhibits · named in 5 exhibits
This roster is in beta and still being tested: every figure is proved against the page it came from, and the way they are read and shown is still changing.
the workshop's gaming laptop: an RTX 5090 Laptop GPU for the diffusion rematch, deliberately unused for the CPU-only bench
It holds — RTX 5090 Laptop GPU (24 GB).
Narrow these 14 figuresevery modelmodel: gemma4:12b · 2 of 14 figuresmodel: gemma4:26b · 2 of 14 figuresmodel: Mistral Small 3.2 · 2 of 14 figuresmodel: nemotron-3.5-lightning:30b-a3b · 2 of 14 figuresmodel: phi4:14b · 2 of 14 figuresmodel: qwen3.5:9b · 2 of 14 figuresmodel: not stated · 2 of 14 figuresevery exhibitexhibit: How the print lab works · 1 of 14 figuresexhibit: Two Hours at 12 tok/s · 13 of 14 figures
The rail is CSS and nothing else — no script runs on this page. It narrows by model and by power cap independently, and the line above says what each one is set to; the size of the two together is the one number it cannot count. A group or a band with nothing left in it goes with its rows, so what you are left looking at is what matched. A narrowed view is not part of the address, so it cannot be linked — each card’s own page is a URL, and that is the thing to share.
Language — decode, prefill and first token
Grouped by model and by what was measured, and inside a group sorted by power cap ascending — alphabetical and numeric, never by value. The heading of a group carries everything that does not change inside it; each row carries the value and the conditions that do. A condition the page does not state reads not stated, which is a third thing and not an uncapped card. A † means the same reading is printed in a second place on its exhibit — one reading, two printings, not two figures.
| value | what moves inside this group |
|---|---|
| gemma4:12b decode, median of five scored runstok/sQ4_K_M · Ollama · 0.32.15 · a 4,096-token context · Ollama's own default of 4 threads · 2026-08-25 (UTC) · power cap and concurrency not stated | |
| 5.0 tok/s | |
from Two Hours at 12 tok/s § The one table to rememberread off that page’s own table by the roster’s reader, then checked against it word for word · the card is named in the table’s own heading · read from two-hours-on-battery · section id #the-one-table-to-remember | |
| gemma4:12b time to first token, median of five cold-prefix runssQ4_K_M · Ollama · 0.32.15 · a 4,096-token context · 536 prompt tokens, Ollama's own default of 4 threads · 2026-08-25 (UTC) · power cap and concurrency not stated | |
| 22.2 s | |
from Two Hours at 12 tok/s § The wait: prefill is the real CPU taxread off that page’s own table by the roster’s reader, then checked against it word for word · the card is named in the table’s own heading · read from two-hours-on-battery · section id #the-wait-prefill-is-the-real-cpu-tax | |
| gemma4:26b decode, median of five scored runstok/sQ4_K_M · Ollama · 0.32.15 · a 4,096-token context · Ollama's own default of 4 threads · 2026-08-25 (UTC) · power cap and concurrency not stated | |
| 13.3 tok/s | |
from Two Hours at 12 tok/s § The one table to rememberread off that page’s own table by the roster’s reader, then checked against it word for word · the card is named in the table’s own heading · read from two-hours-on-battery · section id #the-one-table-to-remember | |
| gemma4:26b the page says: gemma4:26b (MoE) time to first token, median of five cold-prefix runssQ4_K_M · Ollama · 0.32.15 · a 4,096-token context · 536 prompt tokens, Ollama's own default of 4 threads · 2026-08-25 (UTC) · power cap and concurrency not stated | |
| 8.5 s | |
from Two Hours at 12 tok/s § The wait: prefill is the real CPU taxread off that page’s own table by the roster’s reader, then checked against it word for word · the card is named in the table’s own heading · read from two-hours-on-battery · section id #the-wait-prefill-is-the-real-cpu-tax | |
| Mistral Small 3.2 the page says: mistral-small3.2:24b decode, median of five scored runstok/sQ4_K_M · Ollama · 0.32.15 · a 4,096-token context · Ollama's own default of 4 threads · 2026-08-25 (UTC) · power cap and concurrency not stated | |
| 2.7 tok/s | |
from Two Hours at 12 tok/s § The one table to rememberread off that page’s own table by the roster’s reader, then checked against it word for word · the card is named in the table’s own heading · read from two-hours-on-battery · section id #the-one-table-to-remember | |
| Mistral Small 3.2 the page says: mistral-small3.2:24b time to first token, median of five cold-prefix runssQ4_K_M · Ollama · 0.32.15 · a 4,096-token context · 1,023 prompt tokens, Ollama's own default of 4 threads · 2026-08-25 (UTC) · power cap and concurrency not stated | |
| 82.4 s | |
from Two Hours at 12 tok/s § The wait: prefill is the real CPU taxread off that page’s own table by the roster’s reader, then checked against it word for word · the card is named in the table’s own heading · read from two-hours-on-battery · section id #the-wait-prefill-is-the-real-cpu-tax | |
| nemotron-3.5-lightning:30b-a3b decode, median of five scored runstok/sQ4_K_M · Ollama · 0.32.15 · a 4,096-token context · Ollama's own default of 4 threads · 2026-08-25 (UTC) · power cap and concurrency not stated | |
| 15.4 tok/s | |
from Two Hours at 12 tok/s § The one table to rememberread off that page’s own table by the roster’s reader, then checked against it word for word · the card is named in the table’s own heading · read from two-hours-on-battery · section id #the-one-table-to-remember | |
| nemotron-3.5-lightning:30b-a3b the page says: nemotron-3.5-lightning:30b-a3b (MoE) time to first token, median of five cold-prefix runssQ4_K_M · Ollama · 0.32.15 · a 4,096-token context · 532 prompt tokens, Ollama's own default of 4 threads · 2026-08-25 (UTC) · power cap and concurrency not stated | |
| 9.9 s | |
from Two Hours at 12 tok/s § The wait: prefill is the real CPU taxread off that page’s own table by the roster’s reader, then checked against it word for word · the card is named in the table’s own heading · read from two-hours-on-battery · section id #the-wait-prefill-is-the-real-cpu-tax | |
| phi4:14b decode, median of five scored runstok/sQ4_K_M · Ollama · 0.32.15 · a 4,096-token context · Ollama's own default of 4 threads · 2026-08-25 (UTC) · power cap and concurrency not stated | |
| 4.4 tok/s | |
from Two Hours at 12 tok/s § The one table to rememberread off that page’s own table by the roster’s reader, then checked against it word for word · the card is named in the table’s own heading · read from two-hours-on-battery · section id #the-one-table-to-remember | |
| phi4:14b time to first token, median of five cold-prefix runssQ4_K_M · Ollama · 0.32.15 · a 4,096-token context · 505 prompt tokens, Ollama's own default of 4 threads · 2026-08-25 (UTC) · power cap and concurrency not stated | |
| 26.4 s | |
from Two Hours at 12 tok/s § The wait: prefill is the real CPU taxread off that page’s own table by the roster’s reader, then checked against it word for word · the card is named in the table’s own heading · read from two-hours-on-battery · section id #the-wait-prefill-is-the-real-cpu-tax | |
| qwen3.5:9b decode, median of five scored runstok/sQ4_K_M · Ollama · 0.32.15 · a 4,096-token context · Ollama's own default of 4 threads · 2026-08-25 (UTC) · power cap and concurrency not stated | |
| 7.5 tok/s | |
from Two Hours at 12 tok/s § The one table to rememberread off that page’s own table by the roster’s reader, then checked against it word for word · the card is named in the table’s own heading · read from two-hours-on-battery · section id #the-one-table-to-remember | |
| qwen3.5:9b time to first token, median of five cold-prefix runssQ4_K_M · Ollama · 0.32.15 · a 4,096-token context · 537 prompt tokens, Ollama's own default of 4 threads · 2026-08-25 (UTC) · power cap and concurrency not stated | |
| 14.1 s | |
from Two Hours at 12 tok/s § The wait: prefill is the real CPU taxread off that page’s own table by the roster’s reader, then checked against it word for word · the card is named in the table’s own heading · read from two-hours-on-battery · section id #the-wait-prefill-is-the-real-cpu-tax | |
Pictures — seconds per render
1 row — same rule as above.
| value | what moves inside this group |
|---|---|
| not stated the whole board as one sprite, 144 cells144 cells, 128 pixels a cell · 2026-09-09 15:45:32Z to 15:46:47Z · quantisation, runtime, power cap, context and concurrency not stated | |
| 5,191,824 bytes (5.2 MB), 1.12 s | |
from How the print lab works § Numbers, so it isn't just talktyped in from that page by hand, then checked against it word for word · the card is named in the same section · read from how-the-print-lab-works · section id #numbers-so-it-isnt-just-talk | |
Power, temperature, and what it costs
1 row — same rule as above.
| value | what moves inside this group |
|---|---|
| not stated memory bandwidth, best triadGB/sa STREAM triad, gcc -O3 -march=native -fopenmp · version not stated · a 2,289 MiB working set, best of 9 timed reps · 2026-08-25 (UTC), 17:57–19:46Z · quantisation, power cap, context and concurrency not stated | |
| 59.3 GB/s | |
from Two Hours at 12 tok/s § Read this before the table: CPU speed is a memory-bandwidth storyread off that page’s own table by the roster’s reader, then checked against it word for word · the card is named in this row · read from two-hours-on-battery · section id #read-this-before-the-table-cpu-speed-is-a-memory-bandwidth | |
Where it appears
Every released exhibit that names this identity, in the shelf’s own order. A count beside an exhibit is how many of its figures are carried above, and it links the control that narrows the table to them — a number with no way back to what it counts is the thing this shelf exists against.
- A laptop, asked the desktop's questions — named, and no figure carried from it — yet
- A Rig Your Friend Already Owns — named, and no figure carried from it — yet
- How the print lab works — 1 figure on this page
- Six graphics cards, the same questions — named, and no figure carried from it — yet
- Two Hours at 12 tok/s — 13 figures on this page
How this one is named here
The canonical name above is the full product name. Everything below is a spelling a published page or kit has actually used for it, kept so a reader who searches an older page’s words lands on this row. The set is closed and historical: no page cut after 2026-09-17 writes a memory class where it can write a card.
- Rig B — the laptop: 24G-VRAM class, GPU deliberately unused here.
- Rig B laptop
- Rig B †
- cpu-laptop
- test Rig B, a laptop: 24 logical CPUs - 8 performance cores with no hyperthreading plus 16 efficiency cores; hybrid Intel part; a 24G-VRAM-class GPU present and parked
- the workshop's laptop