Speed sweep

pg-01a013a0-0e0a-7e56-9629-019e431c415c-sweep

pg-01a013a0-0e0a-7e56-9629-019e431c415c-sweep

JSON API Back to speed sweep search

Record

Recipe: pg-prism-ml-ternary-bonsai-27b-gguf-q2-0-apple-m5-pro-64gb-llama-cpp-60e4399df3
Measured: 2026-08-18T06:47:31.586Z
Points: 4

Measured speed

Concurrency	Context	Prefill	Decode	TTFT ms	Status
1	8,256	—	25.6	—	observed
1	32,832	—	20.4	—	observed
1	188,352	—	8.9	—	observed
1	32,768	—	—	321,457.7	observed

Remaining fields

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

id: pg-01a013a0-0e0a-7e56-9629-019e431c415c-sweep
measured at: 2026-08-18T06:47:31.586Z
recipe id: pg-prism-ml-ternary-bonsai-27b-gguf-q2-0-apple-m5-pro-64gb-llama-cpp-60e4399df3
schema version: local-ai-registry/v1

metrics

base memory bytes: 26582188032
base memory context tokens: 192
concurrency: 1
decode32k context tokens: 32,832
decode32k tps: 20.362
decode8k context tokens: 8,256
decode8k tps: 25.559
decode max context tokens: 188,352
decode max context tps: 8.926
decode mode: non-mtp
inference engine version: 9591 (62061f910)
latest point at: 2026-08-18T06:47:31.586Z
max context tokens: 188,352
max prompt tokens: 188,288
memory8k bytes: 27873132544
memory8k context tokens: 8,256
memory max context bytes: 32048136192
memory max context tokens: 188,352
peak generation tps: 20.362
peak memory bytes: 32048136192
peak prompt tps: 101.936
point count: 161
ttft32k cached prompt tokens: 28,668
ttft32k context tokens: 32,768
ttft32k seconds: 321.458

rows

concurrency	context tokens	decode tok s	decode tok s per stream	output tokens	peak vram gb	prefill tok s	samples
1	8,256	25.559	25.559	Unknown	25.959	Unknown	1
1	32,832	20.362	20.362	Unknown	29.847	Unknown	1
1	188,352	8.926	8.926	Unknown	29.847	Unknown	1
1	32,768	Unknown	Unknown	Unknown	Unknown	Unknown	1

Provenance & metadata (1)

source

paths: publication:pg-20260827T060320709Z, run:01a013a0-0e0a-7e56-9629-019e431c415c
repository: local.ai ↗