Speed sweep

pg-01a02ade-493b-7998-bcc8-67a2d1cc0f31-sweep

pg-01a02ade-493b-7998-bcc8-67a2d1cc0f31-sweep

JSON API Back to speed sweep search

Record

Recipe: pg-qwen3-6-35b-a3b-iq2m-iq2m-apple-m3-ultra-96gb-60c-llama-cpp-e94dc225d4
Measured: 2026-08-22T19:06:45.938Z
Points: 4

Measured speed

Concurrency	Context	Prefill	Decode	TTFT ms	Status
1	8,256	—	80.1	—	observed
1	32,832	—	70.9	—	observed
1	262,080	—	27.5	—	observed
1	32,768	—	—	28,930	observed

Remaining fields

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

id: pg-01a02ade-493b-7998-bcc8-67a2d1cc0f31-sweep
measured at: 2026-08-22T19:06:45.938Z
recipe id: pg-qwen3-6-35b-a3b-iq2m-iq2m-apple-m3-ultra-96gb-60c-llama-cpp-e94dc225d4
schema version: local-ai-registry/v1

metrics

base memory bytes: 17249124352
base memory context tokens: 192
concurrency: 1
decode32k context tokens: 32,832
decode32k tps: 70.938
decode8k context tokens: 8,256
decode8k tps: 80.101
decode max context tokens: 262,080
decode max context tps: 27.464
decode mode: non-mtp
inference engine version: unknown
latest point at: 2026-08-22T19:06:45.938Z
max context tokens: 262,080
max prompt tokens: 262,016
memory8k bytes: 17760796672
memory8k context tokens: 8,256
memory max context bytes: 20645232640
memory max context tokens: 262,080
peak generation tps: 70.938
peak memory bytes: 20645232640
peak prompt tps: 1,132.666
point count: 224
ttft32k cached prompt tokens: 28,668
ttft32k context tokens: 32,768
ttft32k seconds: 28.93

rows

concurrency	context tokens	decode tok s	decode tok s per stream	output tokens	peak vram gb	prefill tok s	samples
1	8,256	80.101	80.101	Unknown	16.541	Unknown	1
1	32,832	70.938	70.938	Unknown	19.227	Unknown	1
1	262,080	27.464	27.464	Unknown	19.227	Unknown	1
1	32,768	Unknown	Unknown	Unknown	Unknown	Unknown	1

Provenance & metadata (1)

source

paths: publication:pg-20260827T060320709Z, run:01a02ade-493b-7998-bcc8-67a2d1cc0f31
repository: local.ai ↗