decision2:sol
Decision 2.0 · vLLM Semantic Router · 1.88B · ONNX Runtime · Apache-2.0
- Score
- 50.0
- 95 % interval 47.8–52.2
- Rank
- 31
- of 66 scored models
- Public benchmark
- 67.8
- macro accuracy, %
- typed-decisions
- 55.5
- accuracy, %
By task
Chance-corrected accuracy on each of the 14 tasks: 0 is guessing, 100 is every answer right.
Workflows
typed-decisions
34.7
Classification
MASSIVE en
75.5
MASSIVE de
74.0
Civil Comments
60.0
Aegis 2
47.2
Understanding
MultiNLI
85.5
VitaminC
61.7
PAWS
27.2
Knowledge
BoolQ
54.7
SQuAD 2
37.1
PubMedQA
43.6
Judging
HelpSteer 2
24.2
SummEval rel.
7.3
SummEval cons.
67.0
Accuracy and calibration
| Benchmark | Accuracy | Choice | Noul | Score | ECE | Brier | Rejected |
|---|---|---|---|---|---|---|---|
| Public benchmark | 67.8(66.3–69.2) | 75.9 | 72.6 | 46.3 | 0.053 | 0.412 | 0 / 3,880 |
| typed-decisions | 55.5(53.0–57.7) | 49.5 | 67.3 | 51.0 | 0.057 | 0.554 | 0 / 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
| Device | p50 | p90 | Load | VRAM | RAM peak | Precision |
|---|---|---|---|---|---|---|
| RTX 5090 CUDA | 89.4 ms | 104 ms | 44.4 s | 10.8 GB | 11.4 GB | F32 |
| Threadripper 3960X CPU | 4.64 s | 4.74 s | 44.6 s | – | 11.2 GB | F32 |
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
| Device | Same decision | Max logit diff. | Result | Record |
|---|---|---|---|---|
| RTX 5090 CPU | 463/463 | 8.4e-6 | pass | results/runs/2026-10-25-parity-rtx5090-cpu-cob561.json |
| RTX 5090 CUDA | 463/463 | 1.0e-5 | pass | results/runs/2026-10-25-parity-rtx5090-cuda-cob561.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 decision2:sol
# 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 decision2:sol \
--url http://127.0.0.1:11435 --data typed.jsonl --out typed/decision2-sol
# Bespoke Labs' public benchmark; build its 13 subsets first
NIMBLE_SRC=nimble python -m ollaya_convert.bench_public run --model decision2:sol \
--url http://127.0.0.1:11435 --data public --out public/decision2-sol
# speed and memory
python -m ollaya_convert.bench_latency --models decision2:sol --memory \
--machine my-machine --device cuda --out latency.json