Speed sweep

pg-01a036b6-4761-79ac-a95f-d612093c737d-sweep

pg-01a036b6-4761-79ac-a95f-d612093c737d-sweep

JSON API Back to speed sweep search

Record

Recipe: pg-bonsai-27b-q1-0-dspark-q4-1-q1-0-apple-m5-max-128gb-llama-cpp-c8d64ee55f
Measured: 2026-08-25T02:18:30.611Z
Points: 3

Measured speed

Concurrency	Context	Prefill	Decode	TTFT ms	Status
1	8,256	—	58.5	—	observed
1	32,832	—	45.4	154,677.3	observed
1	262,080	—	14.5	—	observed

Remaining fields

Identity, launch, related records, and measured speed are shown above. This is the rest of the normalized record.

id: pg-01a036b6-4761-79ac-a95f-d612093c737d-sweep
measured at: 2026-08-25T02:18:30.611Z
recipe id: pg-bonsai-27b-q1-0-dspark-q4-1-q1-0-apple-m5-max-128gb-llama-cpp-c8d64ee55f
schema version: local-ai-registry/v1

metrics

base memory bytes: 23307649024
base memory context tokens: 192
concurrency: 1
decode32k context tokens: 32,832
decode32k tps: 45.373
decode8k context tokens: 8,256
decode8k tps: 58.501
decode max context tokens: 262,080
decode max context tps: 14.518
decode mode: non-mtp
inference engine version: 9591 (62061f910)
latest point at: 2026-08-25T02:18:30.611Z
max context tokens: 262,080
max prompt tokens: 262,016
memory8k bytes: 24599052288
memory8k context tokens: 8,256
memory max context bytes: 28950282240
memory max context tokens: 262,080
peak generation tps: 45.373
peak memory bytes: 28950282240
peak prompt tps: 211.847
point count: 224
ttft32k cached prompt tokens: 28,668
ttft32k context tokens: 32,768
ttft32k seconds: 154.677

rows

concurrency	context tokens	decode tok s	decode tok s per stream	output tokens	peak vram gb	prefill tok s	samples
1	8,256	58.501	Unknown	Unknown	22.91	Unknown	1
1	32,832	45.373	Unknown	Unknown	26.962	Unknown	1
1	262,080	14.518	Unknown	Unknown	26.962	Unknown	1

Provenance & metadata (1)

source

paths: publication:pg-20260827T060320709Z, run:01a036b6-4761-79ac-a95f-d612093c737d
repository: local.ai ↗