List your company

Assistant Response Generation on WildBench v2

68.4Win RateSFTUSER
30.12840.0645059.936May 10, 2026

Evaluation results

MethodLinks
SFTUSEROpponent=Initial2026.0568.4—
SFTUSEROpponent=RPUSER12026.0563.9—
RPUSER3Opponent=Initial2026.0561.3—
SFTUSEROpponent=RPUSER22026.0560—
RPUSER2Opponent=Initial2026.0559.9—
SFTUSEROpponent=RPUSER32026.0558.5—
RPUSER3Opponent=RPUSER12026.0556.6—
RPUSER1Opponent=Initial2026.0555.4—
RPUSER2Opponent=RPUSER12026.0555.2—
RPUSER3Opponent=RPUSER22026.0551—
RPUSER2Opponent=RPUSER32026.0549—
RPUSER1Opponent=RPUSER22026.0544.8—
Initial assistantOpponent=RPUSER12026.0544.6—
RPUSER1Opponent=RPUSER32026.0543.4—
RPUSER3Opponent=SFTUSER2026.0541.5—
Initial assistantOpponent=RPUSER22026.0540.1—
RPUSER2Opponent=SFTUSER2026.0540—
Initial assistantOpponent=RPUSER32026.0538.7—
RPUSER1Opponent=SFTUSER2026.0536.1—
Initial assistantOpponent=SFTUSER2026.0531.6—
Initial assistant2026.05—51.1
RPUSER12026.05—53.6
RPUSER22026.05—56.6
RPUSER32026.05—57.7
SFTUSER2026.05—61.5