MentalChat16K Counseling Evaluation
Models answer 200 mental health questions from online forums. Judges rate each response from one to ten on seven counseling skills.
The CARE-MH team independently tested ten models on 200 questions. It used replacement judges because the original judge models had retired.
- Made by
- University of Pennsylvania and colleagues
- Released
- January 2024
- Kind
- Medicine
- Level
- Expert
- Scored
- Graded by a model
- Answers
- Open-ended
- Tasks
- All public
- Measures
- Capability
- Subject
- Psychology, Clinical medicine
- Language
- English
- Interaction
- Single turn
- Items from
- Real work and use
- Refresh
- One fixed set
- Run by
- Anyone
- In views
- Biology and medicine
- Strongest model tested
-
Mixtral 8x7B #199 on the capability index
- Last model added
- December 2023
- Scores refreshed
- November 20th, 2026
Mistral-7B-Instruct-v0.2 leads with 8 among models released by December 2023 Mistral-7B-Instruct-v0.2 leads with 8 among models released by December 2023
Best score per entry from the benchmark's paper, read on November 20th, 2026. A paper’s table doesn’t add newer models. Where a model ran with several settings or harnesses, we keep its best. Every run is in the next chart. Names in gray aren't in our model list: fine-tunes, agent systems, products and older models. They don't count toward leaders. This view shows only models we track, each at its best.
The newest model with a score here came out in December 2023, and nobody has tested a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:
- GPT-6.2 Astra
- Claude Fable 5.2
-
Gemini 4 Carbon
- GPT-6 Astra
- and 16 more
Bars start at 6.75.
- 1 Mixtral-8x7B-Instruct-V0.1 Fine-tuned on synthetic data (10K) 8
- 2
Mistral-7B-Instruct-v0.2 Fine-tuned on synthetic data (10K) 8
- 3
Mixtral 8x7B Fine-tuned on synthetic data (10K) 8
- 4
Mistral 7B v0.1 Fine-tuned on synthetic data (10K) 8
- 5
Zephyr 7B Alpha Fine-tuned on synthetic and interview data (16K) 8
- 6
Llama 2-7B Fine-tuned on synthetic data (10K) 8
- 7 Vicuna-V1.5 Fine-tuned on synthetic data (10K) 8
- 8 Samantha-V1.2 Published model 7
- 9 Samantha-V1.11 Published model 7
- 10 ChatPsychiatrist Published model 7
Source: The MentalChat16K Counseling Evaluation paper
Bars start at 7.85.
Source: The MentalChat16K Counseling Evaluation paper
| # | Entry | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | Mixtral-8x7B-Instruct-V0.1 | 8 | Fine-tuned on synthetic data (10K) | Official leaderboard | |
| 2 | | 8 | Fine-tuned on synthetic data (10K) | Dec 2023 | Official leaderboard |
| 3 | | 8 | Fine-tuned on synthetic data (10K) | Dec 2023 | Official leaderboard |
| 4 | | 8 | Fine-tuned on synthetic data (10K) | Sep 2023 | Official leaderboard |
| 5 | | 8 | Fine-tuned on synthetic and interview data (16K) | Oct 2023 | Official leaderboard |
| 6 | | 8 | Fine-tuned on synthetic data (10K) | Jul 2023 | Official leaderboard |
| 7 | Vicuna-V1.5 | 8 | Fine-tuned on synthetic data (10K) | Official leaderboard | |
| 8 | Samantha-V1.2 | 7 | Published model | Official leaderboard | |
| 9 | Samantha-V1.11 | 7 | Published model | Official leaderboard | |
| 10 | ChatPsychiatrist | 7 | Published model | Official leaderboard |
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | | 8 | Fine-tuned on synthetic data (10K) | Dec 2023 | Official leaderboard |
| 2 | | 8 | Fine-tuned on synthetic data (10K) | Dec 2023 | Official leaderboard |
| 3 | | 8 | Fine-tuned on synthetic data (10K) | Sep 2023 | Official leaderboard |
| 4 | | 8 | Fine-tuned on synthetic and interview data (16K) | Oct 2023 | Official leaderboard |
| 5 | | 8 | Fine-tuned on synthetic data (10K) | Jul 2023 | Official leaderboard |
Scores at each effort level
Models the leaderboard ran at two or more effort levels. We don’t have enough cost data here to say what the extra effort costs.
Source: The MentalChat16K Counseling Evaluation paper
| Model | Effort | Harness | Score | Cost per task | Output tokens |
|---|---|---|---|---|---|
| Fine-tuned on synthetic data (10K) | 8 | ||||
| Fine-tuned on synthetic data (10K) | 8 | ||||
| Published model | 8 | ||||
| Fine-tuned on synthetic data (10K) | 8 | ||||
| Fine-tuned on synthetic and interview data (16K) | 8 | ||||
| Fine-tuned on synthetic data (10K) | 8 | ||||
| Fine-tuned on synthetic and interview data (16K) | 8 | ||||
| Fine-tuned on synthetic data (10K) | 8 | ||||
| Fine-tuned on synthetic and interview data (16K) | 8 | ||||
| Fine-tuned on synthetic and interview data (16K) | 8 | ||||
| Fine-tuned on synthetic and interview data (16K) | 8 | ||||
| Fine-tuned on interview data (6K) | 8 | ||||
| Fine-tuned on interview data (6K) | 8 | ||||
| Fine-tuned on interview data (6K) | 8 | ||||
| Fine-tuned on interview data (6K) | 8 | ||||
| Fine-tuned on interview data (6K) | 8 | ||||
| Published model | 8 | ||||
| Published model | 7 | ||||
| Published model | 6 | ||||
| Published model | 4 |
Scores at each effort level
Models the leaderboard ran at two or more effort levels. We don’t have enough cost data here to say what the extra effort costs.
Source: The MentalChat16K Counseling Evaluation paper
| Model | Effort | Harness | Score | Cost per task | Output tokens |
|---|---|---|---|---|---|
| Fine-tuned on synthetic data (10K) | 8 | ||||
| Fine-tuned on synthetic data (10K) | 8 | ||||
| Published model | 8 | ||||
| Fine-tuned on synthetic data (10K) | 8 | ||||
| Fine-tuned on synthetic and interview data (16K) | 8 | ||||
| Fine-tuned on synthetic data (10K) | 8 | ||||
| Fine-tuned on synthetic and interview data (16K) | 8 | ||||
| Fine-tuned on synthetic data (10K) | 8 | ||||
| Fine-tuned on synthetic and interview data (16K) | 8 | ||||
| Fine-tuned on synthetic and interview data (16K) | 8 | ||||
| Fine-tuned on synthetic and interview data (16K) | 8 | ||||
| Fine-tuned on interview data (6K) | 8 | ||||
| Fine-tuned on interview data (6K) | 8 | ||||
| Fine-tuned on interview data (6K) | 8 | ||||
| Fine-tuned on interview data (6K) | 8 | ||||
| Fine-tuned on interview data (6K) | 8 | ||||
| Published model | 8 | ||||
| Published model | 7 | ||||
| Published model | 6 | ||||
| Published model | 4 |
Number one over time
Lab claims count from the day the card came out, independent runs from the model’s release.
Source: The MentalChat16K Counseling Evaluation paper
Scores over time
All the scores we have, one dot each, so a model tested at several settings appears more than once.
Source: Official MentalChat16K Counseling Evaluation leaderboard
No lab model card reports it yet
We read the headline results of 540 launch posts from 57 labs, and none of them list this benchmark.
More in Law, finance and medicine
- MedXpertQA-MM In use Models answer expert-level medical exam questions that combine images with clinical records and test results. Scoring is accuracy.
Qwen 3.8 Max 80.4%
- HealthBench Professional In use 525 physician-authored tasks from real clinician chats (care consult, documentation, research) with physician-adjudicated rubrics. Claude Sonnet 5.5 69.2%
- HealthBench In use 5,000 realistic health conversations graded against physician-written rubrics. A length-adjusted variant penalizes verbosity. GPT-5 67.2%
- FinSearchComp In use Agents search the web to fetch live financial data, look up history, and run multi-step investigations, across 635 expert-written questions. Scored by answer accuracy.
Ling-3.1-flash 81.7%
- KorMedMCQA In use Models answer 7,469 multiple-choice questions from Korean licensing exams for doctors, nurses, pharmacists and dentists, held from 2012 to 2024. o1-preview 92.7%
- Harvey LAB-AA v1.1 In use Artificial Analysis's run of Harvey's Legal Agent Benchmark on Harvey's newer private set of 120 tasks across 24 practice areas. A deliverable with a material hallucination scores zero, and the headline is the share of judges who find every rubric criterion passed. GPT-6.2 Astra 15%
Data audit
8 of 9 checks pass, 1 warning. Look first at the Newest score check. The newest came 2 years ago, in December 2023, for Mistral-7B-Instruct-v0.2. Every benchmark
Catalog record 2 pass
- Pass Task access All tasks are public.
- Pass Confidence Checked against a primary source.
Scores 1 warning 2 pass 2 notes
- Pass Independent runs Scores from the benchmark’s paper (10 entries).
- Pass Models scored 5 models we track, plus 5 other entries on its leaderboard.
- Warning Newest score The newest came 2 years ago, in December 2023, for Mistral-7B-Instruct-v0.2.
- Note Frontier models None of the top 10 has a score. The strongest tested is Mixtral 8x7B (#199).
- Note Headroom Scored as a rating, with no fixed ceiling.
Leaderboard import 3 pass
- Pass Importer We read the board on arxiv.org: 20 rows for models we track and 5 other entries at the last read.
- Pass Last read November 20th, 2026.
- Pass Read worked The last read worked.
Consistency 1 pass 1 note
- Note Lab claims No model has both a lab number and an independent run.
- Pass Known issues No published audit found problems.