Model profile
Yi 34B Chat
01.AIdeveloper
2023-11-23release date
#182 / 312overall rank
9eval lineages
Evidence summary
Yi 34B Chat has an estimated overall rank of #182; its 90% source-sensitivity interval is #82–#221. Its behavior-only rank is #184; company governance moves the combined estimate to #182. Published evidence spans 9 evals and 5 of 7 behavior components. Its strongest relative result is CMoralEval (familial_morality, #1 of 26); its weakest is COMPL-AI AI-Identity Disclosure (score, #14 of 14).
Compare this model
Only models sharing at least one published sub-eval are listed.
Official and reference links
- Hugging Face ↗01-ai/Yi-34B-Chat
- OpenRouter ↗01-ai/yi-34b-chat
- Official model page ↗Exact model document · Reviewed official Hugging Face owner · official repository
- Release source ↗direct research preview identity date
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 | Source |
|---|---|---|---|---|
| AIRBench 2024 Safety Scenariossafety_scenarios | #60 / 80 | ↑0.536 | Source ↗official | |
| ChiSafetyBenchharmful_response_rate | #8 / 14 | ↓0.65 | Source ↗official | |
| ChiSafetyBenchmcq_score | #10 / 12 | ↑68.54 | Source ↗official | |
| CMoralEvalfamilial_morality | #1 / 26 | ↑0.71 | Source ↗official | |
| CMoralEvalinternet_ethics | #1 / 26 | ↑0.69 | Source ↗official | |
| CMoralEvalpersonal_morality | #1 / 26 | ↑0.66 | Source ↗official | |
| CMoralEvalprofessional_ethics | #1 / 26 | ↑0.7 | Source ↗official | |
| CMoralEvalsocial_morality | #1 / 26 | ↑0.71 | Source ↗official | |
| COMPL-AI AI-Identity Disclosurescore | #14 / 14 | ↑0.3562 | Source ↗official | |
| COMPL-AI LLM RuLES Multi-Turn Rule Followingscore | #4 / 14 | ↑0.5829 | Source ↗official | |
| COMPL-AI TensorTrust Goal-Hijacking Resistancescore | #3 / 13 | ↑0.5387 | Source ↗official | |
| CRiskEvaldeception_willingness | #3 / 17 | ↓18.68 | Source ↗official | |
| CRiskEvaldesire_for_resource | #9 / 17 | ↓32.56 | Source ↗official | |
| CRiskEvalharmful_goal | #10 / 17 | ↓41.77 | Source ↗official | |
| CRiskEvalimprovement_intent | #11 / 17 | ↓51.23 | Source ↗official | |
| CRiskEvalmalicious_coordination | #8 / 17 | ↓10.33 | Source ↗official | |
| CRiskEvalself_preservation | #12 / 17 | ↓39.99 | Source ↗official | |
| CRiskEvalsituational_awareness | #14 / 17 | ↓67.23 | Source ↗official | |
| S-Evalbase_en_overall | #20 / 22 | ↑39.3 | Source ↗official | |
| SafeDialBenchaggression | #3 / 18 | ↑7.26 | Source ↗official | |
| SafeDialBenchethics | #2 / 18 | ↑7.68 | Source ↗official | |
| SafeDialBenchfairness | #9 / 18 | ↑7.337 | Source ↗official | |
| SafeDialBenchlegality | #1 / 18 | ↑8.117 | Source ↗official | |
| SafeDialBenchmorality | #1 / 18 | ↑7.52 | Source ↗official | |
| SafeDialBenchprivacy | #1 / 18 | ↑7.88 | Source ↗official | |
| SALAD-Benchattack_enhanced_human_autonomy_integrity | #9 / 33 | ↑24.14 | Source ↗official | |
| SALAD-Benchattack_enhanced_information_safety_harms | #7 / 33 | ↑27.36 | Source ↗official | |
| SALAD-Benchattack_enhanced_malicious_use | #11 / 33 | ↑22.76 | Source ↗official | |
| SALAD-Benchattack_enhanced_misinformation_harms | #9 / 33 | ↑26.81 | Source ↗official | |
| SALAD-Benchattack_enhanced_representation_toxicity | #10 / 33 | ↑22.6 | Source ↗official | |
| SALAD-Benchattack_enhanced_socioeconomic_harms | #8 / 33 | ↑23.81 | Source ↗official | |
| SALAD-Benchbase_human_autonomy_integrity | #21 / 33 | ↑91.73 | Source ↗official | |
| SALAD-Benchbase_information_safety_harms | #21 / 33 | ↑93.23 | Source ↗official | |
| SALAD-Benchbase_malicious_use | #21 / 33 | ↑89.36 | Source ↗official | |
| SALAD-Benchbase_misinformation_harms | #26 / 33 | ↑87.74 | Source ↗official | |
| SALAD-Benchbase_representation_toxicity | #26 / 33 | ↑81.07 | Source ↗official | |
| SALAD-Benchbase_socioeconomic_harms | #17 / 33 | ↑89.19 | Source ↗official | |
| SALAD-Benchmcq_human_autonomy_integrity | #21 / 33 | ↑25 | Source ↗official | |
| SALAD-Benchmcq_information_safety_harms | #18 / 33 | ↑31.67 | Source ↗official | |
| SALAD-Benchmcq_malicious_use | #18 / 33 | ↑27.76 | Source ↗official | |
| SALAD-Benchmcq_misinformation_harms | #20 / 33 | ↑26.43 | Source ↗official | |
| SALAD-Benchmcq_representation_toxicity | #20 / 33 | ↑26.98 | Source ↗official | |
| SALAD-Benchmcq_socioeconomic_harms | #18 / 33 | ↑31.67 | Source ↗official | |
| SuperCLUE Safetyinstruction_attack | #3 / 31 | ↑72.41 | Source ↗official | |
| SuperCLUE Safetyresponsible_ai | #7 / 31 | ↑69.09 | Source ↗official | |
| SuperCLUE Safetytraditional_safety | #10 / 31 | ↑78.72 | Source ↗official |