Fastino's open-weight decision model. A DeBERTa-v3-large encoder that scores label sets supplied at runtime, answers single-label, multi-label, yes/no and ordinal questions in one pass, and extracts spans and relations.
Native MLX FP16 conversion of Laya for Apple silicon. The same ModernBERT-large encoder, decision transformer and heads, running in MLX with no PyTorch; the checkpoint Laya Studio fine-tunes from.
Decides
choice, score, noul, classify, extract, route
choice, score, noul
Architecture
gliner2
laya
Fine-tuned from
fastino/gliner2-large-v1
convai-innovations/laya
License
apache-2.0
apache-2.0
Availability
Open weights + hosted API
Open weights
Hosted by
Fastino
—
Input price
—
—
Decision accuracy
60.2%
—
Calibration error
—
—
Valid action rate
—
—
Median latency
38.3 ms
—
p95 latency
—
—
Evaluation suite
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 gliner2-5-decide and laya-mlx?
gliner2-5-decide is from Fastino Labs and laya-mlx from aac6fef. gliner2-5-decide has open weights and a hosted API; laya-mlx has open weights you can download and run. Both answer choice, score and noul questions. Only gliner2-5-decide answers classify, extract and route. gliner2-5-decide is the smaller model, at 340M parameters to 421M.
Which is more accurate, gliner2-5-decide or laya-mlx?
Only gliner2-5-decide publishes an accuracy figure (60.2% on Fastino fast-decisions suite (17 datasets)); laya-mlx does not, so there is no comparison to make without your own test.
Which is cheaper, gliner2-5-decide or laya-mlx?
gliner2-5-decide: Hosted, price not published, or free to self-host. laya-mlx: Free (open weights). Open weights cost nothing per call beyond your own hardware.
Can I run gliner2-5-decide or laya-mlx locally?
Yes, both: systemone pull fastino-labs/gliner2-5-decide and systemone pull aac6fef/laya-mlx download the weights.
