system-one-bench

~/system-one-bench / compare/fastino-glide-vs-together-ai-tev1-4b-experimental

Head to head

GLiDE vs Tev1 4B Experimental

GLiDE scores 84.1 and Tev1 4B Experimental 71.5 on the System One Index. GLiDE is more accurate on 16 of 16 text datasets, Tev1 4B Experimental on 0. Same cases, same request, same scorer.

Overall

GLiDETev1 4B Experimental
System One Index84.171.5
English83.066.4
Arabic84.875.8
Vision Index82.5–
Calibration error3.6%5.3%
Median latency–232 ms
Cost per 1,000 calls$0.017$0.019

Text datasets, accuracy

GLiDETev1 4B Experimental
AG News English92.091.5
SST-2 English96.094.0
SST-5 English61.543.0
DAIR Emotion English61.556.5
Banking77 English91.51.0
BoolQ English92.091.0
XNLI English English94.787.3
Typed decisions English75.067.0
XNLI French French88.082.7
XNLI Arabic Arabic83.376.0
MASSIVE scenario Arabic73.556.5
MASSIVE intent Arabic91.070.0
OSACT4 offensive Arabic95.084.5
Tweet sentiment Arabic73.570.5
MIRACL rerank Arabic87.085.5
MIRACL pairwise Arabic90.587.5

Image datasets, accuracy

GLiDETev1 4B Experimental
SST-2, as an image94.0–
Arabic sentiment, as an image51.5–
Oxford-IIIT Pet84.5–
VQAv2 yes/no89.0–
CountBench93.5–

Model pages: GLiDE · Tev1 4B Experimental. Latency for API models is the median per call through OpenRouter, timed from a cloud server with 8 calls in flight.