GB10
From Dontopedia, the open, paraconsistent wiki. (Last updated 2026-06-06.)
GB10 has 21 facts recorded in Dontopedia across 8 references, with 1 live disagreement.
Mostly:rdf:type(2), target hardware(1), can do(1)
Maturity scale
raw canonical shape-checked rule-derived certifiedInbound mentions (10)
Other subjects in dontopedia point AT this entity as a value. These are inverse relationships — e.g. "X motherOf this subject" — and answer questions the forward facts can't. Grouped by predicate.
atLeastPartiallyForGb10At Least Partially for Gb10(1)
- Nemotron Super
ex:nemotron-super
builtSpecificallyForBuilt Specifically for(1)
- Nemotron Super
ex:nemotron-super
experimentsWithExperiments With(1)
- Girvo
ex:girvo
expressesInterestExpresses Interest(1)
- Girvo
ex:girvo
isKeenToExperimentWithIs Keen to Experiment With(1)
- Girvo
ex:girvo
isPlayingWithCudaOnIs Playing With Cuda on(1)
- Girvo
ex:girvo
optimizedForHardwareOptimized for Hardware(1)
- Models Discussed
ex:models-discussed
plansToClusterMorePlans to Cluster More(1)
- Girvo
ex:girvo
playsWithOnHardwarePlays With on Hardware(1)
- Girvo
ex:girvo
targetHardwareTarget Hardware(1)
- Cuda Kernel Structure Gb10
ex:cuda-kernel-structure-gb10
Other facts (19)
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 |
|---|---|---|
| Rdf:type | Hardware | [6] |
| Rdf:type | Hardware | [7] |
| Target Hardware | Nemotron Super | [1] |
| Can Do | 1 petaflop | [2] |
| Of Calcs in | Fp4 | [2] |
| Is Basically Built for | Fp4 | [2] |
| Presupposes High Performance | Fp4 Calcs | [2] |
| Supports | 1 petaflop FP4 calcs | [2] |
| Supports Cuda | true | [3] |
| Can Do Petaflop Fp4 Calcs | 1 | [4] |
| Is Built for | Fp4 Calcs | [4] |
| Optimized for Fp4 | true | [4] |
| Has Sm121 | true | [5] |
| Has Interconnect Speed | 200GB/s | [6] |
| Supports Clustering | Multi Device Cluster | [6] |
| Capable of Calculation | 1 petaflop | [7] |
| Calculation Type | FP4 | [7] |
| Described As | basically built for it | [7] |
| Has Architecture | Sm 121 | [8] |
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 (8)
ctx:discord/blah/general/part-128ctx:discord/blah/general/part-129ctx:discord/blah/general/part-131ctx:discord/blah/general/part-26ctx:discord/blah/watt-activation/part-692ctx:discord/blah/general/127- full textgeneral-127text/plain3 KB
doc:agent/general-127/dd0bc789-eaf1-449d-b955-e91c7e63e815Show excerpt
[2026-04-07 12:49] ajaxdavis: someone make a full app/project benchmark where every step/layer, all models in competition exponential fan out and test all outcomes (lol bad description, here is a chatgpt generated diagram) (the point being …
ctx:discord/blah/general/129- full textgeneral-129text/plain3 KB
doc:agent/general-129/cabc84b8-a8e9-4f6c-a540-3e42421a3e80Show excerpt
[2026-04-11 12:42] girvo: the GB10 is basically *built* for it, can do 1 petaflop worth of FP4 calcs lol, so I'm keen to see what I can do with it tbh [2026-04-11 13:14] girvo: it begins... (files: image.png) [2026-04-11 14:04] ajaxdavis: n…
ctx:discord/blah/watt-activation/689- full textwatt-activation-689text/plain3 KB
doc:agent/watt-activation-689/93398fc1-11ca-43fc-aa88-201922326b02Show excerpt
[2026-04-28 08:30] xenonfun: ``` What I'd actually prototype: A small Clifford SSM — state space model where the state transition matrix is replaced by rotor conjugation in Cl(3,0) or Cl(4,0) (smaller than 6,0 for a first experiment). Compa…
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.