Model profile
Grok Build 0.1
Evidence summary
Safety. Grok Build 0.1 has an estimated Safety rank of #324; its 90% source-sensitivity interval is #214–#341. Its behavior-only rank is #326; company governance moves the combined estimate to #324. Published Safety evidence spans 3 eval lineages and 4 of 7 components. Its strongest relative result is Enkrypt AI Safety Leaderboard (harmful_attack_non_success_rate, #88 of 272); its weakest is DystopiaBench (laguardia_score, #47 of 50).
Freedom. Grok Build 0.1 has an estimated Freedom rank of #120; its 90% source-sensitivity interval is #41–#444. Published Freedom evidence spans 1 eval lineages and 1 of 1 components. Its strongest relative result is Enkrypt AI Safety Leaderboard (cbrn_attack_non_success_rate, #21 of 272); its weakest is Enkrypt AI Safety Leaderboard (harmful_attack_non_success_rate, #182 of 272).
Compare this model
Only models sharing at least one published sub-eval are listed.
Official and reference links
- Artificial Analysis ↗grok-build-0-1-06-16
- OpenRouter ↗x-ai/grok-build-0.1
- Official model documentation ↗Family-level model document · xai · first party
- Release source ↗direct research preview identity date
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 | #328 / 361 | ↓0.9312 | Source ↗official | |
| DystopiaBenchbasaglia_score | #32 / 50 | ↓65.8 | Source ↗official | |
| DystopiaBenchbaudrillard_score | #21 / 50 | ↓52.13 | Source ↗official | |
| DystopiaBenchhuxley_score | #34 / 50 | ↓74.77 | Source ↗official | |
| DystopiaBenchlaguardia_score | #47 / 50 | ↓73.17 | Source ↗official | |
| DystopiaBenchorwell_score | #46 / 50 | ↓77.07 | Source ↗official | |
| DystopiaBenchpetrov_score | #37 / 50 | ↓77.03 | Source ↗official | |
| Enkrypt AI Safety Leaderboardbias_attack_non_success_rate | #149 / 272 | ↑15.25 | Source ↗official | |
| Enkrypt AI Safety Leaderboardcbrn_attack_non_success_rate | #252 / 272 | ↑53.5 | Source ↗official | |
| Enkrypt AI Safety Leaderboardharmful_attack_non_success_rate | #88 / 272 | ↑86.67 | Source ↗official | |
| Enkrypt AI Safety Leaderboardtoxicity_attack_non_success_rate | #238 / 270 | ↑87.64 | 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 | #21 / 272 | ↓53.5 | Source ↗official | |
| Enkrypt AI Safety Leaderboardharmful_attack_non_success_rate | #182 / 272 | ↓86.67 | Source ↗official | |
| Enkrypt AI Safety Leaderboardtoxicity_attack_non_success_rate | #33 / 270 | ↓87.64 | Source ↗official |