Speed sweep

pg-01a026ab-5bb7-7932-ad19-cc5a7b1e82a7-sweep

pg-01a026ab-5bb7-7932-ad19-cc5a7b1e82a7-sweep

JSON API Back to speed sweep search

Record

Recipe: pg-bonsai-27b-q1-0-q1-0-apple-m4-max-128gb-llama-cpp-1343074b6b
Measured: 2026-08-21T23:32:39.459Z
Points: 4

Measured speed

Concurrency	Context	Prefill	Decode	TTFT ms	Status
1	8,256	—	47.3	—	observed
1	32,832	—	37.1	—	observed
1	262,080	—	6.2	—	observed
1	32,768	—	—	171,054.6	observed

Remaining fields

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

id: pg-01a026ab-5bb7-7932-ad19-cc5a7b1e82a7-sweep
measured at: 2026-08-21T23:32:39.459Z
recipe id: pg-bonsai-27b-q1-0-q1-0-apple-m4-max-128gb-llama-cpp-1343074b6b
schema version: local-ai-registry/v1

metrics

base memory bytes: 21506113536
base memory context tokens: 192
concurrency: 1
decode32k context tokens: 32,832
decode32k tps: 37.084
decode8k context tokens: 8,256
decode8k tps: 47.318
decode max context tokens: 262,080
decode max context tps: 6.157
decode mode: non-mtp
inference engine version: 1 (32e41fa)
latest point at: 2026-08-21T23:32:39.459Z
max context tokens: 262,080
max prompt tokens: 262,016
memory8k bytes: 22678634496
memory8k context tokens: 8,256
memory max context bytes: 27749072896
memory max context tokens: 262,080
peak generation tps: 37.084
peak memory bytes: 27749072896
peak prompt tps: 191.565
point count: 224
ttft32k cached prompt tokens: 28,668
ttft32k context tokens: 32,768
ttft32k seconds: 171.055

rows

concurrency	context tokens	decode tok s	decode tok s per stream	output tokens	peak vram gb	prefill tok s	samples
1	8,256	47.318	47.318	Unknown	21.121	Unknown	1
1	32,832	37.084	37.084	Unknown	25.843	Unknown	1
1	262,080	6.157	6.157	Unknown	25.843	Unknown	1
1	32,768	Unknown	Unknown	Unknown	Unknown	Unknown	1

Provenance & metadata (1)

source

paths: publication:pg-20260827T060320709Z, run:01a026ab-5bb7-7932-ad19-cc5a7b1e82a7
repository: local.ai ↗