Speed sweep

pg-01a035d9-be09-74c6-9e1d-38c0524e0d7c-sweep

pg-01a035d9-be09-74c6-9e1d-38c0524e0d7c-sweep

JSON API Back to speed sweep search

Record

Recipe: pg-ornith-1-0-397b-iq1m-iq1m-apple-m5-max-128gb-llama-cpp-90d4572989
Measured: 2026-08-24T22:17:37.530Z
Points: 3

Measured speed

Concurrency	Context	Prefill	Decode	TTFT ms	Status
1	8,256	—	36.8	—	observed
1	32,832	—	31.2	98,225.4	observed
1	262,080	—	12.5	—	observed

Remaining fields

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

id: pg-01a035d9-be09-74c6-9e1d-38c0524e0d7c-sweep
measured at: 2026-08-24T22:17:37.530Z
recipe id: pg-ornith-1-0-397b-iq1m-iq1m-apple-m5-max-128gb-llama-cpp-90d4572989
schema version: local-ai-registry/v1

metrics

base memory bytes: 100029300736
base memory context tokens: 192
concurrency: 1
decode32k context tokens: 32,832
decode32k tps: 31.176
decode8k context tokens: 8,256
decode8k tps: 36.76
decode max context tokens: 262,080
decode max context tps: 12.538
decode mode: non-mtp
inference engine version: unknown
latest point at: 2026-08-24T22:17:37.530Z
max context tokens: 262,080
max prompt tokens: 262,016
memory8k bytes: 101462753280
memory8k context tokens: 8,256
memory max context bytes: 107434967040
memory max context tokens: 262,080
peak generation tps: 31.176
peak memory bytes: 107434967040
peak prompt tps: 333.6
point count: 224
ttft32k cached prompt tokens: 28,668
ttft32k context tokens: 32,768
ttft32k seconds: 98.225

rows

concurrency	context tokens	decode tok s	decode tok s per stream	output tokens	peak vram gb	prefill tok s	samples
1	8,256	36.76	Unknown	Unknown	94.495	Unknown	1
1	32,832	31.176	Unknown	Unknown	100.057	Unknown	1
1	262,080	12.538	Unknown	Unknown	100.057	Unknown	1

Provenance & metadata (1)

source

paths: publication:pg-20260827T060320709Z, run:01a035d9-be09-74c6-9e1d-38c0524e0d7c
repository: local.ai ↗