Speed sweep

pg-01a013a0-0c1c-71d9-8cf5-169aa773342f-sweep

pg-01a013a0-0c1c-71d9-8cf5-169aa773342f-sweep

JSON API Back to speed sweep search

Record

Recipe: pg-mlx-community-qwen3-6-35b-a3b-8bit-8bit-apple-m5-max-128gb-mlx-d290886072
Measured: 2026-08-18T06:47:31.093Z
Points: 4

Measured speed

Concurrency	Context	Prefill	Decode	TTFT ms	Status
1	8,256	—	87	—	observed
1	32,832	—	76.6	—	observed
1	262,079	—	37.8	—	observed
1	32,768	—	—	13,246.6	observed

Remaining fields

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

id: pg-01a013a0-0c1c-71d9-8cf5-169aa773342f-sweep
measured at: 2026-08-18T06:47:31.093Z
recipe id: pg-mlx-community-qwen3-6-35b-a3b-8bit-8bit-apple-m5-max-128gb-mlx-d290886072
schema version: local-ai-registry/v1

metrics

base memory bytes: 37134480110
base memory context tokens: 192
concurrency: 1
decode32k context tokens: 32,832
decode32k tps: 76.59
decode8k context tokens: 8,256
decode8k tps: 87.031
decode max context tokens: 262,079
decode max context tps: 37.84
decode mode: non-mtp
inference engine version: 0.31.3
latest point at: 2026-08-18T06:47:31.093Z
max context tokens: 262,079
max prompt tokens: 262,015
memory8k bytes: 39294965790
memory8k context tokens: 8,256
memory max context bytes: 66158983820
memory max context tokens: 262,079
peak generation tps: 76.59
peak memory bytes: 66158983820
peak prompt tps: 2,473.693
point count: 224
ttft32k cached prompt tokens: 28,736
ttft32k context tokens: 32,768
ttft32k seconds: 13.247

rows

concurrency	context tokens	decode tok s	decode tok s per stream	output tokens	peak vram gb	prefill tok s	samples
1	8,256	87.031	87.031	Unknown	36.596	Unknown	1
1	32,832	76.59	76.59	Unknown	61.615	Unknown	1
1	262,079	37.84	37.84	Unknown	61.615	Unknown	1
1	32,768	Unknown	Unknown	Unknown	Unknown	Unknown	1

Provenance & metadata (1)

source

paths: publication:pg-20260827T060320709Z, run:01a013a0-0c1c-71d9-8cf5-169aa773342f
repository: local.ai ↗