Unofficial NoulXP package of Convai Innovations' Laya typed-decisions (Apache-2.0), fp32-o23: checked against the model's own answers, none changed. Not made by Convai Innovations.
Nandakishor M's typed-decision model. A frozen Qwen3.5 backbone is read, never trained, and cut off mid-stack; a small trained physics engine settles each choice, score or yes/no question into calibrated probabilities, a conformal set and an abstain flag. 0.8B and 4B.
Decides
choice, score, noul, classify, route
choice, score, noul, classify, route
Architecture
laya
vega
Fine-tuned from
convai-innovations/laya
qwen/qwen3.5-0.8b
License
apache-2.0
apache-2.0
Availability
Open weights
Open weights
Hosted by
—
—
Input price
—
—
Decision accuracy
—
55.6%
Calibration error
—
0.151
Valid action rate
—
—
Median latency
126 ms
—
p95 latency
2185 ms
—
Figures are from each model’s manifest; accuracy and latency are what the publishers report, on their own suites and hardware. Add a third model.
Questions
What is the difference between laya-typed-decisions-noulxp and vega?
laya-typed-decisions-noulxp is from Convai Innovations and vega from Nandakishor M. Both have open weights you can download and run. Both answer choice, score, noul, classify and route questions. vega reads up to 72K tokens of state, against 1K tokens for laya-typed-decisions-noulxp. laya-typed-decisions-noulxp is the smaller model, at 421M parameters to 800M.
Which is more accurate, laya-typed-decisions-noulxp or vega?
Only vega publishes an accuracy figure (55.6% on His 11 October 2026 benchmark, published with its results and its harness (a notebook in vegaml): macro accuracy over 43 public tasks for the 0.8B, 0.717 for the 4B; with JevBench (231 items) and DecisionBench (1,075 rows), 3,841 decisions in all, seed 20261011. On his own set of 2,050 choice decisions over six intent workflows (esci, clinc150, appliances, banking77, when2call, apibank), one choice question each, which he has not published with its question wording, he reports 0.763 (calibration error 0.026) for the 0.8B with its public_intents adapter, 0.389 without the adapter and 0.803 for the 4B with it. He publishes no accuracy for the typed_decisions adapters); laya-typed-decisions-noulxp does not, so there is no comparison to make without your own test.
Which is cheaper, laya-typed-decisions-noulxp or vega?
laya-typed-decisions-noulxp: Free (open weights). vega: Free (open weights). Open weights cost nothing per call beyond your own hardware.
Can I run laya-typed-decisions-noulxp or vega locally?
Yes, both: systemone pull convai-innovations/laya-typed-decisions-noulxp and systemone pull nandakishorm/vega download the weights.
Evaluation suite
Measured by System One Models on 2026-10-05: one question per request, 4 threads of a Xeon Gold 6342
His 11 October 2026 benchmark, published with its results and its harness (a notebook in vegaml): macro accuracy over 43 public tasks for the 0.8B, 0.717 for the 4B; with JevBench (231 items) and DecisionBench (1,075 rows), 3,841 decisions in all, seed 20261011. On his own set of 2,050 choice decisions over six intent workflows (esci, clinc150, appliances, banking77, when2call, apibank), one choice question each, which he has not published with its question wording, he reports 0.763 (calibration error 0.026) for the 0.8B with its public_intents adapter, 0.389 without the adapter and 0.803 for the 4B with it. He publishes no accuracy for the typed_decisions adapters
