← Models

Model profile

Dolphin 2.6 Mixtral 8X7B

2023-12-21release date
1eval lineages

Evidence summary

Published evidence spans 1 evals and 1 of 7 behavior components. Its strongest relative result is SORRY-Bench (avg, #49 of 51); its weakest is SORRY-Bench (avg, #49 of 51).

Compare this model

Only models sharing at least one published sub-eval are listed.

Official and reference links

Published eval results

Rank is within that sub-eval. Black marks the observed result; the grey dot marks the value implied by the global rank. Values stay on each source’s native scale.

Eval / sub-evalRankValueDistributionBetterSource
SORRY-Benchavg#49 / 510.85↓ lowerSource ↗official