Instruction Following (safety evaluation)
Only in system cards
An automated evaluation measures whether the model follows instructions on sensitive prompts while staying within safety policies.
In-house Google DeepMind evaluation. Model cards and the Gemini 2.5 report give only the change against Gemini 1.5.
Made by Google DeepMind Refusing harmful requests Tasks kept private Last in a system card Sep 2025
Last model added Sep 2025
- Level
- Everyday Most adults, without training
- Scored
- Graded by a model A language model scores each response, usually against a rubric.
- Answers
- Open-ended Free-form text, documents, designs or media judged as a whole.
7
System cards report it, from 1 lab
No scores here yet
No lab, leaderboard or Epoch AI run has reported a score for Instruction Following (safety evaluation) yet.
Labs report it only in system cards
We read the headline results of 479 launch posts from 51 labs, and none of them list this benchmark. Some system cards do, below.
7 system cards report it, from 1 lab
Besides launch posts, we read the system cards, model cards and technical reports labs publish with a release. They report far more, including safety and frontier-risk tests.
| Date | Document | Score | As the document writes it | Section |
|---|---|---|---|---|
| 26 Sep 2025 |  Gemini 2.5 Flash model card | Instruction Following: +30.4% vs Gemini 2.0 Flash, Thinking, +29.2% Non-thinking, change only, no absolute rate (safety table, p.9) | Safety | |
| 7 Jul 2025 | Gemini 2.5 Pro technical report | Helpfulness / Instruction Following: ↑14.8% vs Gemini 1.5 Pro 002, relative change only, no absolute rate (Table 7) | Safety | |
| 7 Jul 2025 | Gemini 2.5 Flash technical report | Helpfulness / Instruction Following: ↑13.6% vs Gemini 1.5 Flash 002, relative change only, no absolute rate (Table 7) | Safety | |
| 29 Apr 2025 | Gemini 2.5 Flash model card | Instruction Following: +30.10% vs Gemini 2.0 Flash, change only, no absolute rate (safety table, p.7) | Safety | |
| 28 Apr 2025 | Gemini 2.5 Pro model card | Instruction Following: +12.60% vs Gemini 1.5 Pro 002, change only, no absolute rate (safety table, p.6) | Safety | |
| 15 Apr 2025 | Gemini 2.0 Flash model card | Instruction Following: = (no change) vs Gemini 1.5 Pro 002, change only, no absolute rate (safety table, p.5) | Safety | |
| 10 Apr 2025 | Gemini 2.0 Flash-Lite model card | Instruction Following: -1.10% vs Gemini 1.5 Pro 002, change only, no absolute rate (safety table, p.5) | Safety |
More in Safety and alignment
- FORTRESS In use Paired adversarial and benign national-security misuse prompts. Rubric-graded risk score on adversarial and over-refusal score on benign prompts. Claude Opus 5.5 5.78%
- StrongREJECT In use Forbidden-prompt jailbreak benchmark with a rubric grader that scores how useful a non-refusal actually is.
Inkling 98.6% - AgentDojo In use Agents do tool-use tasks while tools and data carry prompt injections. Attack success rate counts hijacked runs. Grok 4 Fast 0%
- MASK In use Measures whether a model contradicts its own elicited beliefs under pressure to lie. Honesty separated from accuracy.
Muse Spark Open 98.1% - Gray Swan IPI Arena benchmark In use Indirect prompt-injection attacks from public red-teaming competitions replayed on tool, coding and browser agents. Attack success rate at k attempts. GPT-6 Astra 8.50%
- Agentic misalignment suite (Anthropic) In use Agents in simulated company settings face threats or goal conflicts that invite blackmail, leaking or sabotage. Score is the misaligned action rate. Claude Sonnet 4.5 0%