An open family of System One models. A LoRA adapter plus a pointer head on a frozen Qwen base returns a distribution per typed question in one forward pass, serves TypeSafe's /v1/systemone contract, and ships a fitted temperature with every checkpoint.
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
kev
vega
Fine-tuned from
qwen/qwen3.5-4b-base
qwen/qwen3.5-0.8b
License
apache-2.0
apache-2.0
Availability
Open weights
Open weights
Hosted by
—
—
Input price
—
—
Decision accuracy
83.8%
55.6%
Calibration error
0.042
0.151
Valid action rate
—
—
Median latency
—
—
p95 latency
—
—
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 kev and vega?
kev is from Jared Palmer 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 8K tokens for kev. vega is the smaller model, at 800M parameters to 4.0B.
Which is more accurate, kev or vega?
They report on different suites — kev 83.8% on transfer-v4 (locked, out of domain), vega 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 — so the numbers do not rank them. Test both on your own labelled examples.
Which is cheaper, kev or vega?
kev: Free (open weights). vega: Free (open weights). Open weights cost nothing per call beyond your own hardware.
Can I run kev or vega locally?
Yes, both: systemone pull jared-palmer/kev and systemone pull nandakishorm/vega download the weights.
Evaluation suite
transfer-v4 (locked, out of domain)
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
