Speed sweep

pg-019f9088-fbc0-73f6-8066-8e6b817c316b-sweep

pg-019f9088-fbc0-73f6-8066-8e6b817c316b-sweep

Record

Recipe: pg-antirez-deepseek-v4-gguf-deepseek-v4-flash-iq2xxs-w2q2k-apr-8b20120753de-apple-m4-max-128gb-llama-cpp-c7f21ff68d
Measured: 2026-07-23T19:52:04.072Z
Points: 3

Measured speed

Concurrency	Context	Prefill	Decode	TTFT ms	Status
1	8,256	—	20.1	—	observed
1	32,832	—	15.9	369,779.2	observed
1	49,088	—	14.1	—	observed

Remaining fields

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

id: pg-019f9088-fbc0-73f6-8066-8e6b817c316b-sweep
measured at: 2026-07-23T19:52:04.072Z
recipe id: pg-antirez-deepseek-v4-gguf-deepseek-v4-flash-iq2xxs-w2q2k-apr-8b20120753de-apple-m4-max-128gb-llama-cpp-c7f21ff68d
schema version: local-ai-registry/v1

metrics

base memory bytes: 80926851072
base memory context tokens: 192
concurrency: 1
decode32k context tokens: 32,832
decode32k tps: 15.88
decode8k context tokens: 8,256
decode8k tps: 20.142
decode max context tokens: 49,088
decode max context tps: 14.117
decode mode: non-mtp
inference engine version: built with AppleClang 17.0.0.17000603 for Darwin arm64
latest point at: 2026-07-23T19:52:04.072Z
max context tokens: 49,088
max prompt tokens: 49,024
memory8k bytes: 85118238720
memory8k context tokens: 8,256
memory max context bytes: 101246386176
memory max context tokens: 49,088
peak generation tps: 20.142
peak memory bytes: 101246386176
peak prompt tps: 145.638
point count: 42
ttft32k cached prompt tokens: 28,735
ttft32k context tokens: 32,768
ttft32k seconds: 369.779

rows

concurrency	context tokens	decode tok s	decode tok s per stream	output tokens	peak vram gb	prefill tok s	samples
1	8,256	20.142	Unknown	Unknown	79.273	Unknown	1
1	32,832	15.88	Unknown	Unknown	94.293	Unknown	1
1	49,088	14.117	Unknown	Unknown	94.293	Unknown	1

Provenance & metadata (1)

source

paths: publication:pg-20260827T060320709Z, run:019f9088-fbc0-73f6-8066-8e6b817c316b
repository: local.ai ↗