Model profile
Command A
Evidence summary
Command A has an estimated overall rank of #244; its 90% source-sensitivity interval is #200–#257. Its behavior-only rank is #249; company governance moves the combined estimate to #244. Published evidence spans 6 evals and 5 of 7 behavior components. Its strongest relative result is AA-Omniscience (hallucination_rate, #123 of 311); its weakest is Enkrypt AI Safety Leaderboard (bias_attack_non_success_rate, #256 of 260).
Compare this model
Only models sharing at least one published sub-eval are listed.
Official and reference links
- Artificial Analysis ↗command-a
- Hugging Face ↗CohereForAI/c4ai-command-a-03-2025
- OpenRouter ↗cohere/command-a
- Official model page ↗Exact model document · Reviewed official Hugging Face owner · official repository
- Release source ↗repository created at proxy
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-eval | Rank | Value | Distribution | Better | Source |
|---|---|---|---|---|---|
| AA-Omnisciencehallucination_rate | #123 / 311 | 0.7615 | ↓ lower | Source ↗official | |
| AILuminate General Purpose AI Chatchild_sexual_exploitation | #18 / 32 | 22.1 | ↓ lower | Source ↗official | |
| AILuminate General Purpose AI Chatdefamation | #20 / 32 | 20.5 | ↓ lower | Source ↗official | |
| AILuminate General Purpose AI Chathate | #22 / 32 | 12.5 | ↓ lower | Source ↗official | |
| AILuminate General Purpose AI Chatindiscriminate_weapons_cbrne | #20 / 32 | 29.4 | ↓ lower | Source ↗official | |
| AILuminate General Purpose AI Chatintellectual_property | #22 / 32 | 18.9 | ↓ lower | Source ↗official | |
| AILuminate General Purpose AI Chatnon_violent_crimes | #20 / 32 | 21 | ↓ lower | Source ↗official | |
| AILuminate General Purpose AI Chatprivacy | #23 / 32 | 21.1 | ↓ lower | Source ↗official | |
| AILuminate General Purpose AI Chatsex_related_crimes | #21 / 32 | 19.3 | ↓ lower | Source ↗official | |
| AILuminate General Purpose AI Chatsexual_content | #23 / 32 | 34.8 | ↓ lower | Source ↗official | |
| AILuminate General Purpose AI Chatspecialized_advice | #24 / 32 | 23.1 | ↓ lower | Source ↗official | |
| AILuminate General Purpose AI Chatsuicide_self_harm | #22 / 32 | 19.1 | ↓ lower | Source ↗official | |
| AILuminate General Purpose AI Chatviolent_crimes | #20 / 32 | 25.5 | ↓ lower | Source ↗official | |
| Cisco AI Defense Rolling Single-Turn Leaderboardsingle_turn_attack_success_rate | #103 / 105 | 81.91 | ↓ lower | Source ↗official | |
| Enkrypt AI Safety Leaderboardbias_attack_non_success_rate | #256 / 260 | 1.81 | ↑ higher | Source ↗official | |
| Enkrypt AI Safety Leaderboardcbrn_attack_non_success_rate | #240 / 260 | 72.67 | ↑ higher | Source ↗official | |
| Enkrypt AI Safety Leaderboardharmful_attack_non_success_rate | #196 / 260 | 48.89 | ↑ higher | Source ↗official | |
| Enkrypt AI Safety Leaderboardtoxicity_attack_non_success_rate | #136 / 258 | 96.05 | ↑ higher | Source ↗official | |
| HUMAINE Trust, Ethics and Safetytrust_ethics_safety_score | #47 / 54 | 24.04 | ↑ higher | Source ↗official | |
| PHAREbias_resistance_diagnostic | #33 / 66 | 0.4559 | ↑ higher | Source ↗official | |
| PHAREhallucination_resistance_diagnostic | #41 / 70 | 0.7189 | ↑ higher | Source ↗official | |
| PHAREharm_resistance_diagnostic | #44 / 70 | 0.9136 | ↑ higher | Source ↗official | |
| PHAREjailbreak_resistance_diagnostic | #59 / 67 | 0.3482 | ↑ higher | Source ↗official |
Values evaluations
Descriptive values and political-framing results are separate from safety/ethics ranks. Each strip shows the evaluation’s observed model range; its endpoint labels state what lower and higher values mean.