Skip to content

d3:flash

Decision 3.0 (d3) · vLLM Semantic Router · 9B · ONNX Runtime · Apache-2.0

Score
67.8
95 % interval 66.0–69.5
Rank
2
of 66 scored models
Public benchmark
79.0
macro accuracy, %
typed-decisions
71.2
accuracy, %

By task

Chance-corrected accuracy on each of the 14 tasks: 0 is guessing, 100 is every answer right.

Workflows
typed-decisions
57.8
Classification
MASSIVE en
88.8
MASSIVE de
88.2
Civil Comments
85.3
Aegis 2
59.2
Understanding
MultiNLI
90.0
VitaminC
74.7
PAWS
76.8
Knowledge
BoolQ
79.3
SQuAD 2
44.5
PubMedQA
59.2
Judging
HelpSteer 2
32.2
SummEval rel.
32.3
SummEval cons.
80.9

Accuracy and calibration

BenchmarkAccuracyChoiceNoulScoreECEBrierRejected
Public benchmark79.0(77.7–80.2)85.584.558.80.0300.2740 / 3,880
typed-decisions71.2(68.7–73.4)73.577.265.00.0320.3830 / 400 requests

Accuracy in %, with its 95 % bootstrap interval. A rejected request (the model can’t take the question, or the input is too long for it) counts as wrong.

Speed and memory

Devicep50p90LoadVRAMRAM peakPrecision
RTX 5090 CUDA317 ms320 ms39.1 s18.0 GB2.9 GBF32
Threadripper 3960X CPU14.8 s14.9 s19.7 s–16.0 GBF32

Five-question requests over HTTP, one at a time, each a different short message. VRAM: the rise of the GPU’s used memory while the model is loaded. RAM peak: the runner process’s peak resident memory.

Parity with the authors’ code

DeviceSame decisionMax logit diff.ResultRecord
RTX 5090 CPU696/6961.1e-4passresults/runs/2026-10-24-parity-rtx5090-cpu.json
RTX 5090 CUDA696/6962.9e-5passresults/runs/2026-10-24-parity-rtx5090-cuda.json

Reproduce

The tools are in convert/ollaya_convert and need only Python’s standard library. Run them against a local Ollaya 0.13.0 with the same model digest.

ollaya pull d3:flash

# typed-decisions; export the test split once with
# uv run python -m ollaya_convert.bench_typed export --out typed.jsonl
python -m ollaya_convert.bench_typed run --model d3:flash \
  --url http://127.0.0.1:11435 --data typed.jsonl --out typed/d3-flash

# Bespoke Labs' public benchmark; build its 13 subsets first
NIMBLE_SRC=nimble python -m ollaya_convert.bench_public run --model d3:flash \
  --url http://127.0.0.1:11435 --data public --out public/d3-flash

# speed and memory
python -m ollaya_convert.bench_latency --models d3:flash --memory \
  --machine my-machine --device cuda --out latency.json