[KongchangAI]
Benchmark

S1MB

A public leaderboard for evaluating decision engines or models, comprising 152 models.

Timeline (last 90 days)

Oct 14

S1MB uses the Borda count method, ranking multiple specialized benchmarks separately and then converting them into scores to average, avoiding a single task dominating the final result.

Unverified50%
Oct 14

Darwin-27B-ZTC-v5 took first place on the S1MB Japanese track with a Borda score of 84.57, leading second-place Perplexity (80.86) by 3.71 points.

Unverified50%
Oct 14

Darwin ZTC is also first place on the S1MB English track (160 models), with a score of 90.80.

Unverified50%
Oct 14

S1MB (System One Mosaic Benchmark) is a benchmark specifically for evaluating decision models, which output judgments directly rather than generating text.

Unverified50%
Oct 14

The operator of the S1MB leaderboard added a Japanese track, japanese-v1, containing 27 Japanese benchmarks.

Unverified50%
Oct 13

VIDRAFT's judgment engine model Darwin-27B-ZTC-v5 ranked first among 152 models on the public benchmark S1MB, and the company says its performance exceeds the models of Perplexity, Cloudflare, and Meta.

Unverified50%
Oct 13

On the public S1MB leaderboard, Darwin-27B-ZTC-v5 ranks first among 152 models.

Unverified50%

All Facts (7)

Source Articles