Speed sweep

pg-01a02848-3ab2-7dca-8eb4-1a0abad9c1bc-sweep

pg-01a02848-3ab2-7dca-8eb4-1a0abad9c1bc-sweep

JSON API Back to speed sweep search

Record

Recipe: pg-mimo-v2-flash-iq1s-iq1s-apple-m3-ultra-96gb-80c-llama-cpp-781616228f
Measured: 2026-08-22T07:03:37.378Z
Points: 4

Measured speed

Concurrency	Context	Prefill	Decode	TTFT ms	Status
1	8,256	—	50.8	—	observed
1	32,832	—	45.7	—	observed
1	262,080	—	13.5	—	observed
1	32,768	—	—	84,936.4	observed

Remaining fields

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

id: pg-01a02848-3ab2-7dca-8eb4-1a0abad9c1bc-sweep
measured at: 2026-08-22T07:03:37.378Z
recipe id: pg-mimo-v2-flash-iq1s-iq1s-apple-m3-ultra-96gb-80c-llama-cpp-781616228f
schema version: local-ai-registry/v1

metrics

base memory bytes: 93721280512
base memory context tokens: 192
concurrency: 1
decode32k context tokens: 32,832
decode32k tps: 45.663
decode8k context tokens: 8,256
decode8k tps: 50.757
decode max context tokens: 262,080
decode max context tps: 13.46
decode mode: non-mtp
inference engine version: unknown
latest point at: 2026-08-22T07:03:37.378Z
max context tokens: 262,080
max prompt tokens: 262,016
memory8k bytes: 93971677184
memory8k context tokens: 8,256
memory max context bytes: 95770624000
memory max context tokens: 262,080
peak generation tps: 48.753
peak memory bytes: 95770624000
peak prompt tps: 454.345
point count: 224
ttft32k cached prompt tokens: 28,680
ttft32k context tokens: 32,768
ttft32k seconds: 84.936

rows

concurrency	context tokens	decode tok s	decode tok s per stream	output tokens	peak vram gb	prefill tok s	samples
1	8,256	50.757	50.757	Unknown	87.518	Unknown	1
1	32,832	45.663	45.663	Unknown	89.193	Unknown	1
1	262,080	13.46	13.46	Unknown	89.193	Unknown	1
1	32,768	Unknown	Unknown	Unknown	Unknown	Unknown	1

Provenance & metadata (1)

source

paths: publication:pg-20260827T060320709Z, run:01a02848-3ab2-7dca-8eb4-1a0abad9c1bc
repository: local.ai ↗