Speed sweep

pg-01a026ab-507c-70bf-86a7-9890df9993e5-sweep

pg-01a026ab-507c-70bf-86a7-9890df9993e5-sweep

JSON API Back to speed sweep search

Record

Recipe: pg-laguna-s-2-1-q4km-unknown-apple-m3-ultra-96gb-60c-llama-cpp-e9261c7942
Measured: 2026-08-21T23:32:36.590Z
Points: 3

Measured speed

Concurrency	Context	Prefill	Decode	TTFT ms	Status
1	8,256	—	47.8	—	observed
1	32,832	—	42.1	63,000.2	observed
1	65,600	—	36.5	—	observed

Remaining fields

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

id: pg-01a026ab-507c-70bf-86a7-9890df9993e5-sweep
measured at: 2026-08-21T23:32:36.590Z
recipe id: pg-laguna-s-2-1-q4km-unknown-apple-m3-ultra-96gb-60c-llama-cpp-e9261c7942
schema version: local-ai-registry/v1

metrics

base memory bytes: 77832994816
base memory context tokens: 192
concurrency: 1
decode32k context tokens: 32,832
decode32k tps: 42.094
decode8k context tokens: 8,256
decode8k tps: 47.823
decode max context tokens: 65,600
decode max context tps: 36.514
decode mode: non-mtp
inference engine version: 10017 (7dc1bfead)
latest point at: 2026-08-21T23:32:36.590Z
max context tokens: 65,600
max prompt tokens: 65,536
memory8k bytes: 78128709632
memory8k context tokens: 8,256
memory max context bytes: 78797127680
memory max context tokens: 65,600
peak generation tps: 39.204
peak memory bytes: 78797127680
peak prompt tps: 441.722
point count: 56
ttft32k cached prompt tokens: 28,735
ttft32k context tokens: 32,768
ttft32k seconds: 63

rows

concurrency	context tokens	decode tok s	decode tok s per stream	output tokens	peak vram gb	prefill tok s	samples
1	8,256	47.823	Unknown	Unknown	72.763	Unknown	1
1	32,832	42.094	Unknown	Unknown	73.386	Unknown	1
1	65,600	36.514	Unknown	Unknown	73.386	Unknown	1

Provenance & metadata (1)

source

paths: publication:pg-20260827T060320709Z, run:01a026ab-507c-70bf-86a7-9890df9993e5
repository: local.ai ↗