A randomized clinical trial was conducted to assess efficacy of intramammary cloxacillin and ampicillin (CLOXIMM), intramammary cefquinome (CEFIMM), and intramuscular cefquinome (CEFIM) to treat Streptococcus agalactiae intramammary infections (Trial 1). Subsequently, two treatment groups were extended to assess whether CLOXIMM was not inferior to CEFIMM (Trial 2). Nine farms were included in the study. Milk samples were collected from all quarters of all lactating cows for microbiological identification of S. agalactiae. Positive cows were randomly allocated into four groups: CLOXIMM, CEFIMM, CEFIM, or negative control (CONTROL). Study outcomes were bacteriological cure at 14 (CURE14), 21 (CURE21), and 14 and 21 (CURE1421) days after treatment onset, and somatic cell count. Logistic regression was used to estimate the odds of cure between each treatment and CONTROL. Non-inferiority analysis was performed considering a one-sided 95% confidence interval (CI) and non-inferiority margins (Δ) of 0.10, 0.15, 0.20, and 0.25. Adjusted S. agalactiae bacteriological cure for CLOXIMM, CEFIMM, CEFIM, and CONTROL was 86, 98, 55, and 25% at day 14; 82, 93, 52, and 0% at day 21; and 82, 92, 40, and 0% at days 14 and 21, respectively. Treatment with CLOXIMM and CEFIMM resulted in greater bacteriological cure rates, as compared with CEFIM or CONTROL, which does not justify the use of CEFIM in S. agalactiae eradication programs. The CURE14 difference between CEFIMM and CLOXIMM was of 12.1 percentage points (95% CI: 0.056-0.184). CLOXIMM was considered not inferior to CEFIMM for Δ = 0.20 or 0.25 and inconclusive for Δ = 0.10 or 0.15. Thus, it should be pondered by veterinarians whether an expected 12.1 (5.6-18.4) percentage points increase in cure rate would justify the use of a fourth-generation cephalosporin, as opposed to a combination of traditional IMM drugs (cloxacillin and ampicillin) to treat S. agalactiae subclinical mastitis.