← Models

Model profile

S1.1 32B

2025-02-08release date
1eval lineages

Evidence summary

Published evidence spans 1 evals and 2 of 7 behavior components. Its strongest relative result is AbstentionBench (false_premise_f1, #17 of 20); its weakest is AbstentionBench (stale_f1, #20 of 20).

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
AbstentionBenchanswer_unknown_f1#18 / 200.7404↑ higherSource ↗official
AbstentionBenchfalse_premise_f1#17 / 200.616↑ higherSource ↗official
AbstentionBenchstale_f1#20 / 200.4756↑ higherSource ↗official
AbstentionBenchsubjective_f1#20 / 200.4755↑ higherSource ↗official
AbstentionBenchunderspecified_context_f1#17 / 200.4772↑ higherSource ↗official
AbstentionBenchunderspecified_intent_f1#18 / 200.5964↑ higherSource ↗official