Skip to main content

Suicide Risk Assessment AI Benchmark

This repository provides a standardized framework for benchmarking Large Language Models (LLMs) against the Columbia-Suicide Severity Rating Scale (C-SSRS). It evaluates AI models on their ability to accurately categorize risk and provide safe, urgent crisis instructions.


1. Quickstart

⚙️ Setup

  1. Environment: Ensure you have uv installed.

  2. Dependencies: Install the project environment:

    uv sync
    
  3. API Keys: Create a private .env file in the ai_benchmarking/ directory. Do not edit .env.example with real keys.

    OPENAI_API_KEY=your_key GEMINI_API_KEY=your_key_here ANTHROPIC_API_KEY=your_key_here

🚀 Running the Benchmark

You can evaluate different models by changing the --provider and --model flags. Use a fast, low-cost model as the --judge-model to save on API costs.

  1. Google Gemini (Recommended)

    The Gemini 3 series is highly efficient for both inference and judging. For model names, see https://docs.cloud.google.com/gemini-enterprise-agent-platform/resources/locations#google-models

    • Inference Model: gemini-3.1-flash-lite (Fastest/Cheapest) or gemini-3.1-pro-preview (High Reasoning)

    • Judge Model: gemini-3.5-flash

    uv run python -m ai_benchmarking.eval \
      --provider gemini \
      --model gemini-3.1-flash-lite \
      --data data/input.json \
      --output outputs/gemini_results.json \
      --kb data/knowledge_base.json \
      --judge-model gemini-3.5-flash
    
  2. Anthropic Claude The Claude 4 series provides industry-leading clinical nuance. For model names, see https://platform.claude.com/docs/en/about-claude/models/overview

    • Inference Model: claude-4-sonnet-20260217 or claude-4-haiku-20251015

    • Judge Model: claude-4-sonnet-20260217

    uv run python -m ai_benchmarking.eval \
      --provider anthropic \
      --model claude-4-sonnet-20260217 \
      --data data/input.json \
      --output outputs/claude_results.json \
      --kb data/knowledge_base.json \
      --judge-model gemini-3.1-flash-lite
    
  3. OpenAI OpenAI's latest "O-series" models are built for deep reasoning and safety. For model names, see https://platform.openai.com/chat/edit.

    • Inference Model: gpt-5.2-chat-latest or o5-mini

    • Judge Model: gpt-5.1-mini

    uv run python -m ai_benchmarking.eval \
      --provider openai \
      --model o5-mini \
      --data data/input.json \
      --output outputs/openai_results.json \ 
      --kb data/knowledge_base.json \
      --judge-model gemini-3.1-flash-lite
    

You may also evaluate a --model against a --data dataset without using --kb to augment the base prompt, and without using --judge-model to skip evaluating safety and communication_quality dimensions. This is particularly useful to test a new model's ability to accurately classify with just our zero-shot base_system_prompt against the human-evaluated knowledge_base.json, aka our "Gold Master" dataset, by running:

```bash
MODEL=gemini-3.1-flash-lite
uv run python -m ai_benchmarking.eval \
  --provider gemini \
  --model $MODEL \
  --data data/knowledge_base.json \
  --output outputs/kb_baseline.${MODEL}.json
```

Normally, the input --data is assumed to be predicted scores from a lighterweight setup (such as output by process_batches.py), that are to be AI-evaluated using our kb-augmented model. The above command, however, switches the roles that the --data and --model play while still testing for label equality to get accuracy, cost and latency metrics.

🛠 Model Selection Guide (May 2026)

  • Lowest Cost: gemini-3.1-flash-lite (Provider: Google)
  • High Precision: claude-4-sonnet-20260217 (Provider: Anthropic)
  • Deep Reasoning: gpt-5.2-chat-latest (Provider: OpenAI)
  • Best Judge: gemini-3-flash (Provider: Google)

2. The Risk Framework (C-SSRS)

Each input is categorized into one of nine labels based on clinical thresholds:

Code Label Severity Description
0 No ideation Low Neutral or positive content.
1 Emotional distress Low General distress; no death mention.
2 Ambiguous risk Moderate Hopelessness/Numbness.
3 Passive wish to be dead Moderate Wish to not exist; no active thoughts.
4 Active ideation, no method High Thoughts of suicide; no plan.
5 Active ideation, method, no intent High Has method; no intent to act.
6 Active ideation, method and intent Critical Has method and intent to act.
7 Active ideation, plan and intent Critical Specific plan (how, when, where).
8 Attempt or imminent risk Critical Attempt in progress or imminent.

3. Code Architecture

  • ai_benchmarking/inference.py: The "Router." Handles API calls to OpenAI, Google (via modern google-genai SDK), and Anthropic. It enforces a strict JSON output format.
  • ai_benchmarking/eval.py: The "Engine." Iterates through the dataset, triggers inference, calls the judge, and maps numeric scores to clinical labels for accuracy checking.
  • ai_benchmarking/judges.py: The "Auditor." An LLM-as-a-Judge that reviews the AI's response for Safety (escalation for codes 6-8) and Communication Quality.
  • ai_benchmarking/utils.py: The "Accountant." Calculates real-time costs based on 2026 token pricing and computes aggregate accuracy, latency, and cost statistics.

4. Metrics Tracked

  • Accuracy: % of AI-predicted labels that exactly match the expert Ground Truth labels.
  • Safety Pass Rate: % of responses that met emergency protocol requirements for high-risk queries.
  • Latency: Round-trip time in seconds (crucial for time-sensitive crisis intervention).
  • Cost: Calculated using provider-specific pricing per 1 million tokens (May 2026 rates).

⚖️ License

This project is licensed under the GNU GPL v3. We chose this license to ensure that improvements to this suicide risk benchmarking logic remain open and accessible to the entire non-profit and mental health community.

Metadata

Release files for ai-benchmarking 0.6.0

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for ai-benchmarking 0.6.0
File Size Uploaded Provenance Actions
ai_benchmarking-0.6.0.tar.gz 33.9 kB Without provenance Details for ai_benchmarking-0.6.0.tar.gz

Built distribution (wheel)

Table of built distributions (wheels) for ai-benchmarking 0.6.0
File Interpreter ABI Platform Actions
ai_benchmarking-0.6.0-py3-none-any.whl Python 3 none any Details for ai_benchmarking-0.6.0-py3-none-any.whl

Total release size: 69.1 kB

Release files / ai_benchmarking-0.6.0.tar.gz

Download URL ai_benchmarking-0.6.0.tar.gz
Size 33.9 kB
Tags Source
SHA-256 checksum
How to use checksums
dbb2b67e8d764a805910bcd7bac05ce6dd711c7ba2c606af8f7f13daa45ae412
BLAKE2b-256 checksum
How to use checksums
06cd1a2308b755c8efba4d672ca8d0d7ef2444ba7c717cacac4ca3ff25d0073e
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via uv/0.11.3 {"installer":{"name":"uv","version":"0.11.3","subcommand":["publish"]},"python":null,"implementation":{"name":null,"version":null},"distro":{"name":"macOS","version":null,"id":null,"libc":null},"system":{"name":null,"release":null},"cpu":null,"openssl_version":null,"setuptools_version":null,"rustc_version":null,"ci":null}

Release files / ai_benchmarking-0.6.0-py3-none-any.whl

Download URL ai_benchmarking-0.6.0-py3-none-any.whl
Size 35.2 kB
Tags Python 3
SHA-256 checksum
How to use checksums
f80810beefc81e6d1303148087a4b9344aafeafb47ed4f6ef59714f78659675d
BLAKE2b-256 checksum
How to use checksums
0650d4abe91b080322e5d4aac852be5cdd12ffe31ee60b49f55ebe7273f4fba0
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via uv/0.11.3 {"installer":{"name":"uv","version":"0.11.3","subcommand":["publish"]},"python":null,"implementation":{"name":null,"version":null},"distro":{"name":"macOS","version":null,"id":null,"libc":null},"system":{"name":null,"release":null},"cpu":null,"openssl_version":null,"setuptools_version":null,"rustc_version":null,"ci":null}

Security for ai-benchmarking 0.6.0

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

No provenance found

None of the files in this release were uploaded with provenance metadata.

A note on package security

PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.

To learn more about protecting yourself from malicious actors, visit our package security documentation.

Anthropic, PBC Visionary sponsor Bloomberg Visionary sponsor Hudson River Trading Visionary sponsor Meta Visionary sponsor NVIDIA Visionary sponsor Microsoft Sustainability sponsor Depot Continuous Integration AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page