Speed sweep

phi-4-mini-instruct-gguf-q4-k-m-intel-arc-pro-b70-32gb-llama-cpp-tp1-sweep

phi-4-mini-instruct-gguf-q4-k-m-intel-arc-pro-b70-32gb-llama-cpp-tp1-sweep

Record

Recipe
phi-4-mini-instruct-gguf-q4-k-m-intel-arc-pro-b70-32gb-llama-cpp-tp1
Measured
2026-07-04T22:45:49.730Z
Points
1

Measured speed

ConcurrencyContextPrefillDecodeTTFT msStatus
14,09696.569.9observed

Remaining fields

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

id
phi-4-mini-instruct-gguf-q4-k-m-intel-arc-pro-b70-32gb-llama-cpp-tp1-sweep
measured at
2026-07-04T22:45:49.730Z
recipe id
phi-4-mini-instruct-gguf-q4-k-m-intel-arc-pro-b70-32gb-llama-cpp-tp1
schema version
local-ai-registry/v1

metrics

concurrency
1
inference engine version
fdb1db877
latest point at
2026-07-04T22:45:49.730Z
max context tokens
4,096
peak generation tps
96.548
point count
1

rows

concurrencycontext tokensdecode tok sdecode tok s per streamoutput tokenspeak vram gbprefill tok ssamples
14,09696.54896.548128UnknownUnknown1
Provenance & metadata (1)