Page values for "GLM 5.3 Flash on 4x 8 GB/Confirmation 6"
From CMP Wiki
More actions
"Confirmation" values
1 row is stored for this page| Field | Field type | Allowed values | Value |
|---|---|---|---|
| page | Page | GLM 5.3 Flash on 4x 8 GB | |
| confirmed_by | String | darkmaniac7 | |
| date | Date | 2026-10-19 | |
| cards | String | 8 GB · 10 GB · both | 8 GB |
| result | String | Reproduced · Partly reproduced · Did not reproduce | Reproduced |
| numbers | Text | Recipe 1.8.0 as published, MiaAI protocol, Gen2 x8, no P2P, 74 SMs, 180 W: PP4 one user 201 / 149 / 97 tok/s (structured / code / prose), 664 / 548 / 477 at eight users, 32k cold prefill 6,834-6,977 tok/s; TP4 361 / 284 / 199, 697 / 587 / 529 at eight users, prefill 1,633-1,753. On an 8-card box at Gen2 x16 with P2P (abliterated W4A16 checkpoint, own bench with 1,024-token answers, code / structured / prose): TP4 252 / 235 / 139, 623 / 617 / 421 at eight users, prefill 1.7k. | |
| notes | Wikitext | PP4 reproduces the published numbers closely; TP4 lands at about 0.75x without x16 links and P2P, because the all-reduce runs through host memory. Draft acceptance matched the published regime. Vision and video worked. ?'"`UNIQ--templatestyles-00000006-QINU`"'?Source: @darkmaniac7, 2026-10-19 (lab notes) |