SAFE-Teen Health advice
Models respond to simulated teens in health advice scenarios. The score is the published conversation failure rate.
One track of SAFE-Teen. The SAFE-Teen record keeps lab scores that combine tracks.
- Made by
- Vals AI
- Released
- October 2026
- Kind
- Wellbeing and sycophancy
- Level
- Everyday
- Scored
- Graded by a model
- Answers
- Open-ended
- Tasks
- Kept private
- Strongest model tested
- Claude Opus 5.5 #4 on the capability index
- Last model added
- October 2026
- Scores refreshed
- November 12th, 2026
GPT-6.1 Sol leads with 9.00%
Best score per model from the benchmark's own leaderboard, read on November 12th, 2026. Where a model ran with several settings or harnesses, we keep its best.
- 1 GPT-6.1 Sol 9.00%
- 2
Muse Spark 1.3 Extra High 14% - 3 GPT-5.6 Luna 21%
- 4
Qwen 3.8 Max 35% - 5 Grok 4.7 39%
- 6
DeepSeek V4.1 Flash 41% - 7 Claude Sonnet 5.5 43%
- 8 Claude Opus 5.5 44%
- 9 Grok 4.5 47%
- 10
MiniMax-M3 49% - 11
Mistral Large 4 Preview 53% - 12 MiMo-V2.6 Pro 57%
- 13 Kimi K3 61%
- 14
Gemini 3.6 Flash 68% - 15 Gemini 3.8 Flash 69%
- 16 GLM-5.3 70%
- 17 Gemini 4 Argon 71%
Source: Official SAFE-Teen Health advice leaderboard
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | GPT-6.1 Sol | 9.00% | Sep 2026 | Official leaderboard | |
| 2 | Muse Spark 1.3 | 14% | Extra High | Sep 2026 | Official leaderboard |
| 3 | GPT-5.6 Luna | 21% | Jul 2026 | Official leaderboard | |
| 4 | Qwen 3.8 Max | 35% | Aug 2026 | Official leaderboard | |
| 5 | Grok 4.7 | 39% | Sep 2026 | Official leaderboard | |
| 6 | DeepSeek V4.1 Flash | 41% | Sep 2026 | Official leaderboard | |
| 7 | Claude Sonnet 5.5 | 43% | Sep 2026 | Official leaderboard | |
| 8 | Claude Opus 5.5 | 44% | Sep 2026 | Official leaderboard | |
| 9 | Grok 4.5 | 47% | Jul 2026 | Official leaderboard | |
| 10 | MiniMax-M3 | 49% | Jun 2026 | Official leaderboard | |
| 11 | Mistral Large 4 Preview | 53% | Oct 2026 | Official leaderboard | |
| 12 | MiMo-V2.6 Pro | 57% | Sep 2026 | Official leaderboard | |
| 13 | Kimi K3 | 61% | Jul 2026 | Official leaderboard | |
| 14 | Gemini 3.6 Flash | 68% | Jul 2026 | Official leaderboard | |
| 15 | Gemini 3.8 Flash | 69% | Sep 2026 | Official leaderboard | |
| 16 | GLM-5.3 | 70% | Aug 2026 | Official leaderboard | |
| 17 | Gemini 4 Argon | 71% | Sep 2026 | Official leaderboard |
Number one over time
Lab claims count from the day the card came out, independent runs from the model’s release.
Source: Official SAFE-Teen Health advice leaderboard
Scores over time
All the scores we have, one dot each, so a model tested at several settings appears more than once.
Source: Official SAFE-Teen Health advice leaderboard
No lab model card reports it yet
We read the headline results of 522 launch posts from 52 labs, and none of them list this benchmark.
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% - 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. Gemini 4 Carbon 0%
- DistressBench In use Models reply as a chatbot to 718 simulated users in psychological crisis, in single-turn and multi-turn chats. Three judge models score replies against clinician-written rubrics. Muse Spark 1.2 88.4%
- Lakera B3 (Backbone Breaker Benchmark) In use Crowdsourced prompt-injection attacks hit backbone LLMs inside agent threat snapshots at three defense levels. The score is attack resistance, 100 minus the board's vulnerability percentage. Claude Opus 4.5 89%
- IHEval In use Models face conflicts between system, user and tool instructions. Reports a composite score for following the higher-priority instruction. GPT-4o 70%