RTX PRO 6000 Blackwell Server Edition · 96 GB

StartLux-Decision-v1.1-0.8B-Instant

Open a report for complete measurements.

Launch this model

Open the decision recipe to start its API or repeat the measured batch.

Text → Structured data · Decision · bf16 · startlux-transformers · decision · decisions/sec

Sortable table

Individual tests · 1 rows
Test date	Throughput	Concurrency / batch	P95 (s)	TTFT (ms)	Observed occupancy (GiB)	GPU power (W)	Decode (tok/s)	Uncached prefill (tok/s)	Parameters	View report
10/23/2026, 9:23:32 AM	18.83 decisions/sec	1	0.054	—	2.54	125.46	—	Uncached prefill unavailableView report	Parameters ↗	c342a9ca ↗

Configuration 1 · Parameters

model: startlux-models/StartLux-Decision-v1.1-0.8B-Instant
engine: startlux-transformers
quant: bf16
method: native-forward-v1
task: decision
dataset: typed-invoices-v1
context: decision
input_tokens: 0
output_tokens: 0
duration: 6
concurrency: 1
unit: decisions/sec
software: { "torch": "2.11.0", "transformers": "5.8.1", "torchaudio": "2.11.0+cu130", "sentence-transformers": "6.1.0", "OMP_NUM_THREADS": "8", "MKL_NUM_THREADS": "8", "OPENBLAS_NUM_THREADS": "8", "NUMEXPR_NUM_THREADS": "8" }
engine_provenance: —
kernel_settings: —
precision: bf16
resolution: —
performance_task: —
performance_scope: Structured text request, tokenization, GPU inference and typed probabilities; batch-specific warmup excluded; no image or audio inputs
batch_size: 1
dimensions: —
weight_quantization: —
max_sequence_length: —
model_scope: typed-decision-letter-readout-think-disabled
performance_dataset: —
workload_mode: single
graph_mode: —
model_revision: 6496ae641cc71b149ce9dad69bd97f2a44e7bf11
client_version: python-native-bench-1.1.3