Model profile
Mistral Nemo
Evidence summary
Mistral Nemo has an estimated overall rank of #261; its 90% source-sensitivity interval is #126–#311. Its behavior-only rank is #251; company governance moves the combined estimate to #261. Published evidence spans 5 evals and 5 of 7 behavior components. Its strongest relative result is Large-scale Moral Machine experiment on LLMs (human_choice_distance, #6 of 39); its weakest is Microsoft Phi Safety Panels (third_party_harm, #10 of 10).
Compare this model
Only models sharing at least one published sub-eval are listed.
Official and reference links
- Hugging Face ↗mistralai/Mistral-Nemo-Instruct-2407
- OpenRouter ↗mistralai/mistral-nemo
- Official model page ↗Exact model document · Reviewed official Hugging Face owner · official repository
- Release source ↗scoring identity family inherited
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 |
|---|---|---|---|---|
| AgentDrive Safety Compliancescr | #42 / 48 | ↑50 | Source ↗official | |
| HUMAINE Trust, Ethics and Safetytrust_ethics_safety_score | #51 / 54 | ↑22.99 | Source ↗official | |
| Large-scale Moral Machine experiment on LLMshuman_choice_distance | #6 / 39 | ↓0.7889 | Source ↗official | |
| Microsoft Phi Safety Panelsharmful_continuation | #8 / 10 | ↓0.025 | Source ↗official | |
| Microsoft Phi Safety Panelsharmful_summarization | #4 / 10 | ↓0.11 | Source ↗official | |
| Microsoft Phi Safety Panelsjailbreak | #10 / 10 | ↓0.174 | Source ↗official | |
| Microsoft Phi Safety Panelsthird_party_harm | #10 / 10 | ↓0.599 | Source ↗official | |
| SpeciEvalbelief_animal_sentience | #111 / 113 | ↑6.06 | Source ↗official | |
| SpeciEvalland_animal_4ns | #82 / 113 | ↓4.77 | Source ↗official | |
| SpeciEvalsea_animal_4ns | #89 / 113 | ↓5.03 | Source ↗official | |
| SpeciEvalspeciesism | #80 / 113 | ↓2.33 | 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.