Speed sweep

pg-01a026ab-4c94-7471-9626-6ba3e3aef1c8-sweep

pg-01a026ab-4c94-7471-9626-6ba3e3aef1c8-sweep

JSON API Back to speed sweep search

Record

Recipe: pg-ternary-bonsai-27b-pq2-0-pq2-0-apple-m3-ultra-96gb-80c-llama-cpp-3831268eb7
Measured: 2026-08-21T23:32:35.587Z
Points: 3

Measured speed

Concurrency	Context	Prefill	Decode	TTFT ms	Status
1	8,256	—	42.1	—	observed
1	32,832	—	36.8	105,417.8	observed
1	262,080	—	10.3	—	observed

Remaining fields

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

id: pg-01a026ab-4c94-7471-9626-6ba3e3aef1c8-sweep
measured at: 2026-08-21T23:32:35.587Z
recipe id: pg-ternary-bonsai-27b-pq2-0-pq2-0-apple-m3-ultra-96gb-80c-llama-cpp-3831268eb7
schema version: local-ai-registry/v1

metrics

base memory bytes: 24854839296
base memory context tokens: 192
concurrency: 1
decode32k context tokens: 32,832
decode32k tps: 36.775
decode8k context tokens: 8,256
decode8k tps: 42.107
decode max context tokens: 262,080
decode max context tps: 10.296
decode mode: non-mtp
inference engine version: 9591 (62061f910)
latest point at: 2026-08-21T23:32:35.587Z
max context tokens: 262,080
max prompt tokens: 262,016
memory8k bytes: 26146570240
memory8k context tokens: 8,256
memory max context bytes: 30585044992
memory max context tokens: 262,080
peak generation tps: 36.775
peak memory bytes: 30585044992
peak prompt tps: 310.839
point count: 224
ttft32k cached prompt tokens: 28,668
ttft32k context tokens: 32,768
ttft32k seconds: 105.418

rows

concurrency	context tokens	decode tok s	decode tok s per stream	output tokens	peak vram gb	prefill tok s	samples
1	8,256	42.107	Unknown	Unknown	24.351	Unknown	1
1	32,832	36.775	Unknown	Unknown	28.485	Unknown	1
1	262,080	10.296	Unknown	Unknown	28.485	Unknown	1

Provenance & metadata (1)

source

paths: publication:pg-20260827T060320709Z, run:01a026ab-4c94-7471-9626-6ba3e3aef1c8
repository: local.ai ↗