Speed sweep

qwen3-30b-a3b-instruct-2507-gguf-ud-q4-k-xl-intel-arc-pro-b70-32gb-llama-cpp-tp1-sweep

qwen3-30b-a3b-instruct-2507-gguf-ud-q4-k-xl-intel-arc-pro-b70-32gb-llama-cpp-tp1-sweep

Record

Recipe: qwen3-30b-a3b-instruct-2507-gguf-ud-q4-k-xl-intel-arc-pro-b70-32gb-llama-cpp-tp1
Measured: 2026-07-04T19:42:22.927Z
Points: 1

Measured speed

Concurrency	Context	Prefill	Decode	TTFT ms	Status
1	4,096	—	107.5	167	observed

Remaining fields

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

id: qwen3-30b-a3b-instruct-2507-gguf-ud-q4-k-xl-intel-arc-pro-b70-32gb-llama-cpp-tp1-sweep
measured at: 2026-07-04T19:42:22.927Z
recipe id: qwen3-30b-a3b-instruct-2507-gguf-ud-q4-k-xl-intel-arc-pro-b70-32gb-llama-cpp-tp1
schema version: local-ai-registry/v1

metrics

concurrency: 1
inference engine version: llama-server version 9763 (dec5ca557), IntelLLVM 2026.0.0; source snapshot fdb1db877c526ec90f668eca1b858da5dba85560
latest point at: 2026-07-04T19:42:22.927Z
max context tokens: 4,096
peak generation tps: 107.484
point count: 1

rows

concurrency	context tokens	decode tok s	decode tok s per stream	output tokens	peak vram gb	prefill tok s	samples
1	4,096	107.484	107.484	128	Unknown	Unknown	1

Provenance & metadata (1)

source