system-one-bench

~/system-one-bench / models/fastino-glide

Fastino · Fastino API

GLiDE

GLiDE by Fastino scores 84.1 on the System One Index (3rd of 39) and 82.5 on the Vision Index (8th of 14). Calibration error 3.6%.

System One Index
84.1
3rd of 39
English
83.0
Arabic
84.8
Vision Index
82.5
8th of 14
Calibration error
3.6%
Lower is better
Median latency
–
Cost per 1,000 calls
$0.017
At list price

Text datasets

DatasetAccuracyRankBest on the boardCalibration error
AG News English92.02nd of 3992.56.2%
SST-2 English96.05th of 3998.03.5%
SST-5 English61.53rd of 3962.57.2%
DAIR Emotion English61.511th of 3991.511.3%
Banking77 English91.59th of 3998.53.8%
BoolQ English92.03rd of 3993.02.8%
XNLI English English94.72nd of 3995.34.2%
Typed decisions English75.06th of 3980.06.3%
XNLI French French88.06th of 3991.33.9%
XNLI Arabic Arabic83.35th of 4190.02.9%
MASSIVE scenario Arabic73.59th of 4189.513.6%
MASSIVE intent Arabic91.03rd of 4191.52.9%
OSACT4 offensive Arabic95.04th of 4197.03.3%
Tweet sentiment Arabic73.58th of 4180.05.7%
MIRACL rerank Arabic87.017th of 4190.55.4%
MIRACL pairwise Arabic90.59th of 4191.52.4%

Image datasets

DatasetAccuracyRankBest on the boardCalibration error
SST-2, as an image94.08th of 1497.05.1%
Arabic sentiment, as an image51.59th of 1475.013.0%
Oxford-IIIT Pet84.510th of 1498.08.5%
VQAv2 yes/no89.03rd of 1489.52.7%
CountBench93.53rd of 1495.53.2%

This run

Model id
fastino/GLiDE
Runs on
Fastino API
Run date
2026-11-02
Latency
API; not timed (round 10 ran on Fastino's own API from a machine whose round trips don't compare with the board's)
Questions
4,850, 0 failed (counted wrong)

Ran on Fastino's own API, which reports the model as glide_v2. The open weights are V1.2 (fastino/GLiDE, 26B, Fastino's community licence). Cost is input tokens at the $0.04 per 1M list price, since the API reports no cost. Doesn't answer Box questions.

Compare

Ranks count the models charted on the leaderboard. Every model gets the same cases and the same request, under the benchmark rules.