Speed sweep

pg-01a02a01-9ef5-7254-8910-9da1b95f0a8e-sweep

pg-01a02a01-9ef5-7254-8910-9da1b95f0a8e-sweep

JSON API Back to speed sweep search

Record

Recipe: pg-qwen3-6-35b-a3b-q4km-unknown-apple-m4-pro-48gb-llama-cpp-5a3180cb8f
Measured: 2026-08-22T15:05:44.190Z
Points: 4

Measured speed

Concurrency	Context	Prefill	Decode	TTFT ms	Status
1	8,256	—	46.2	—	observed
1	32,832	—	38.6	—	observed
1	262,080	—	7.3	—	observed
1	32,768	—	—	88,478.9	observed

Remaining fields

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

id: pg-01a02a01-9ef5-7254-8910-9da1b95f0a8e-sweep
measured at: 2026-08-22T15:05:44.190Z
recipe id: pg-qwen3-6-35b-a3b-q4km-unknown-apple-m4-pro-48gb-llama-cpp-5a3180cb8f
schema version: local-ai-registry/v1

metrics

base memory bytes: 27852226560
base memory context tokens: 192
concurrency: 1
decode32k context tokens: 32,832
decode32k tps: 38.57
decode8k context tokens: 8,256
decode8k tps: 46.179
decode max context tokens: 262,080
decode max context tps: 7.294
decode mode: non-mtp
inference engine version: unknown
latest point at: 2026-08-22T15:05:44.190Z
max context tokens: 262,080
max prompt tokens: 262,016
memory8k bytes: 28364423168
memory8k context tokens: 8,256
memory max context bytes: 31229558784
memory max context tokens: 262,080
peak generation tps: 38.57
peak memory bytes: 31229558784
peak prompt tps: 370.348
point count: 224
ttft32k cached prompt tokens: 28,668
ttft32k context tokens: 32,768
ttft32k seconds: 88.479

rows

concurrency	context tokens	decode tok s	decode tok s per stream	output tokens	peak vram gb	prefill tok s	samples
1	8,256	46.179	46.179	Unknown	26.416	Unknown	1
1	32,832	38.57	38.57	Unknown	29.085	Unknown	1
1	262,080	7.294	7.294	Unknown	29.085	Unknown	1
1	32,768	Unknown	Unknown	Unknown	Unknown	Unknown	1

Provenance & metadata (1)

source

paths: publication:pg-20260827T060320709Z, run:01a02a01-9ef5-7254-8910-9da1b95f0a8e
repository: local.ai ↗