Speed sweep

llama-3-1-70b-instruct-q4-k-m-ryzen-ai-max-plus-395-128gb-llama-cpp-tp1-sweep

llama-3-1-70b-instruct-q4-k-m-ryzen-ai-max-plus-395-128gb-llama-cpp-tp1-sweep

Record

Recipe
llama-3-1-70b-instruct-q4-k-m-ryzen-ai-max-plus-395-128gb-llama-cpp-tp1
Measured
2026-04-26T15:05:53.642Z
Points
1

Measured speed

ConcurrencyContextPrefillDecodeTTFT msStatus
116,38449.35.2observed

Remaining fields

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

id
llama-3-1-70b-instruct-q4-k-m-ryzen-ai-max-plus-395-128gb-llama-cpp-tp1-sweep
measured at
2026-04-26T15:05:53.642Z
recipe id
llama-3-1-70b-instruct-q4-k-m-ryzen-ai-max-plus-395-128gb-llama-cpp-tp1
schema version
local-ai-registry/v1

metrics

concurrency
1
inference engine version
b8672
latest point at
2026-04-26T15:05:53.642Z
max context tokens
16,384
peak generation tps
5.18
peak prompt tps
49.33
point count
1

rows

concurrencycontext tokensdecode tok sdecode tok s per streamoutput tokenspeak vram gbprefill tok ssamples
116,3845.185.18128Unknown49.331
Provenance & metadata (1)