GPU decode pipeline
From Dontopedia, the open, paraconsistent wiki. (Last updated 2026-06-06.)
GPU decode pipeline has 13 facts recorded in Dontopedia across 2 references.
Mostly:uses metal kernels(1), dispatches kernels per layer(1), has number of layers(1)
Maturity scale
raw canonical shape-checked rule-derived certifiedOther facts (12)
The long tail: predicates that appear too rarely to warrant their own section. Filter or scroll to find a specific one. Each row links to its source.
| Predicate | Value | Ref |
|---|---|---|
| Uses Metal Kernels | Metal Kernel | [1] |
| Dispatches Kernels Per Layer | 8 | [1] |
| Has Number of Layers | 8 | [1] |
| Includes Classifier | 1 | [1] |
| Presupposes Metal Framework | Metal | [1] |
| Total Dispatches | 65 | [1] |
| Causes Dispatch Overhead | Dispatch | [1] |
| Rdf:type | Pipeline | [2] |
| Dispatches Kernel Count | 8 | [2] |
| Has Layer Count | 8 | [2] |
| Has Classifier Count | 1 | [2] |
| Total Dispatch Count | 65 | [2] |
Timeline
Timeline axis is valid_time — when each source says the fact was true in the world, not when Dontopedia learned about it. Retracted rows are kept for provenance; coloured stripes indicate the context kind.
References (2)
ctx:discord/blah/resources/part-46ctx:discord/blah/resources/46- full textresources-46text/plain3 KB
doc:agent/resources-46/c161ccd3-a97a-462d-a712-6bf0ddc94ec8Show excerpt
[2026-03-03 01:04] xenonfun: going to do a further optimization pass: ``` The GPU decode pipeline dispatches 8 Metal kernels per layer × 8 layers + 1 classifier = 65 dispatches. Each dispatch has overhead (~1-3μs for dispatch + memory barr…
See also
Keep researching
Missing something or suspicious of what's here? Kick off a research session — a Claude agent will investigate, cite its sources, and file new facts into a dedicated context you can review before accepting into the shared view.