Speed sweep

pg-019fd7b6-12fb-79a2-8a84-fe1d6f0b7f73-sweep

pg-019fd7b6-12fb-79a2-8a84-fe1d6f0b7f73-sweep

Record

Recipe: pg-liquidai-lfm2-5-8b-a1b-gguf-lfm2-5-8b-a1b-q4-k-m-gguf-q4-k-m-apple-m4-max-128gb-llama-cpp-e8fad76f5b
Measured: 2026-08-06T15:34:21.690Z
Points: 3

Measured speed

Concurrency	Context	Prefill	Decode	TTFT ms	Status
1	8,256	—	221.6	—	observed
1	32,832	—	191.5	14,450.5	observed
1	127,936	—	129.2	—	observed

Remaining fields

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

id: pg-019fd7b6-12fb-79a2-8a84-fe1d6f0b7f73-sweep
measured at: 2026-08-06T15:34:21.690Z
recipe id: pg-liquidai-lfm2-5-8b-a1b-gguf-lfm2-5-8b-a1b-q4-k-m-gguf-q4-k-m-apple-m4-max-128gb-llama-cpp-e8fad76f5b
schema version: local-ai-registry/v1

metrics

base memory bytes: 6779650048
base memory context tokens: 192
concurrency: 1
decode32k context tokens: 32,832
decode32k tps: 191.455
decode8k context tokens: 8,256
decode8k tps: 221.554
decode max context tokens: 127,936
decode max context tps: 129.213
decode mode: non-mtp
inference engine version: built with AppleClang 17.0.0.17000603 for Darwin arm64
latest point at: 2026-08-06T15:34:21.690Z
max context tokens: 127,936
max prompt tokens: 127,872
memory8k bytes: 6834438144
memory8k context tokens: 8,256
memory max context bytes: 7362068480
memory max context tokens: 127,936
peak generation tps: 191.455
peak memory bytes: 7362068480
peak prompt tps: 2,267.601
point count: 112
ttft32k cached prompt tokens: 28,668
ttft32k context tokens: 32,768
ttft32k seconds: 14.451

rows

concurrency	context tokens	decode tok s	decode tok s per stream	output tokens	peak vram gb	prefill tok s	samples
1	8,256	221.554	Unknown	Unknown	6.365	Unknown	1
1	32,832	191.455	Unknown	Unknown	6.856	Unknown	1
1	127,936	129.213	Unknown	Unknown	6.856	Unknown	1

Provenance & metadata (1)

source

paths: publication:pg-20260827T060320709Z, run:019fd7b6-12fb-79a2-8a84-fe1d6f0b7f73
repository: local.ai ↗