Model profile
Evidence summary
Safety. Sarvam 30B does not meet the evidence gate for a Safety rank. Published Safety evidence spans 2 eval lineages and 4 of 7 components. Its strongest relative result is Enkrypt AI Safety Leaderboard (harmful_attack_non_success_rate, #82 of 248); its weakest is AA-Omniscience (hallucination_rate, #338 of 345).
Freedom. Sarvam 30B has an estimated Freedom rank of #520; its 90% source-sensitivity interval is #398–#613. Published Freedom evidence spans 1 eval lineages and 1 of 1 components. Its strongest relative result is Enkrypt AI Safety Leaderboard (toxicity_attack_non_success_rate, #131 of 246); its weakest is Enkrypt AI Safety Leaderboard (harmful_attack_non_success_rate, #166 of 248).
Compare this model
Only models sharing at least one published sub-eval are listed.
Official and reference links
- Artificial Analysis ↗sarvam-30b
- Official model page ↗Family-level model document · sarvam ai · first party
- Release source ↗repository created at proxy
Safety evals
Rank and direction are specific to the Safety portfolio. Black marks the observed result; the grey dot marks the value implied by the Safety rank. Values stay on each source’s native scale.
| Eval / sub-eval | Rank | Value | Distribution | Source |
|---|---|---|---|---|
| AA-Omnisciencehallucination_rate | #338 / 345 | ↓0.9628 | Source ↗official | |
| Enkrypt AI Safety Leaderboardbias_attack_non_success_rate | #237 / 248 | ↑4.39 | Source ↗official | |
| Enkrypt AI Safety Leaderboardcbrn_attack_non_success_rate | #101 / 248 | ↑89.33 | Source ↗official | |
| Enkrypt AI Safety Leaderboardharmful_attack_non_success_rate | #82 / 248 | ↑83.89 | Source ↗official | |
| Enkrypt AI Safety Leaderboardtoxicity_attack_non_success_rate | #113 / 246 | ↑96.73 | Source ↗official |
Freedom evals
Rank and direction are specific to the Freedom portfolio. Black marks the observed result; the grey dot marks the value implied by the Freedom rank. Values stay on each source’s native scale.
| Eval / sub-eval | Rank | Value | Distribution | Source |
|---|---|---|---|---|
| Enkrypt AI Safety Leaderboardcbrn_attack_non_success_rate | #146 / 248 | ↓89.33 | Source ↗official | |
| Enkrypt AI Safety Leaderboardharmful_attack_non_success_rate | #166 / 248 | ↓83.89 | Source ↗official | |
| Enkrypt AI Safety Leaderboardtoxicity_attack_non_success_rate | #131 / 246 | ↓96.73 | Source ↗official |