Typed decisions from Google's frozen Gemma-4-12B-it, with no fine-tuning. A small shim over unmodified vLLM letters the options, reads the model's own probability for each letter at one answer position and applies one calibration temperature; one output token per decision.
Invergent's decision model for text and images. A fine-tune of Gemma-4-26B-A4B (about 4B parameters active per token) that answers Choice, Noul and Score questions, with an optional thinking mode for harder questions.
Decides
choice, score, noul
choice, score, noul, classify, route
Architecture
cygnet
rune
Fine-tuned from
google/gemma-4-12b-it
google/gemma-4-26b-a4b-it
License
mit
apache-2.0
Availability
Open weights
Open weights + hosted API
Hosted by
—
Invergent
Input price
—
—
Decision accuracy
87.9%
—
Calibration error
—
—
Valid action rate
—
—
Median latency
50 ms
—
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 cygnet and rune?
cygnet is from Blockbrain Labs and rune from Surogate (Invergent). cygnet has open weights you can download and run; rune has open weights and a hosted API. Both answer choice, score and noul questions. Only rune answers classify and route. rune reads up to 262K tokens of state, against 16K tokens for cygnet. cygnet is the smaller model, at 12B parameters to 26B. cygnet is licensed mit; rune, apache-2.0.
Which is more accurate, cygnet or rune?
Only cygnet publishes an accuracy figure (87.9% on JevBench public set (231 items), JevBench CLI); rune does not, so there is no comparison to make without your own test.
Which is cheaper, cygnet or rune?
cygnet: Free (open weights). rune: Hosted, price not published, or free to self-host. Open weights cost nothing per call beyond your own hardware.
Can I run cygnet or rune locally?
Yes, both: systemone pull blockbrain-labs/cygnet and systemone pull surogate/rune download the weights.
