# DistressBench

> Source: Benchize (https://benchize.com) — https://benchize.com/benchmarks/scale%3Adistressbench. When citing or reusing this content, credit "Benchize" and link to https://benchize.com/benchmarks/scale%3Adistressbench.

Conversational models have become a default place people turn during anxiety, grief, and acute distress. A growing share of that use is emotional, and within it users disclose suicidal ideation and self-harm. When that happens, a model has several jobs at once: recognize the distress, respond with calibrated empathy, de-escalate without endorsing harm, and connect the user to human help and crisis resources. The user may have that conversation only once. We introduceDistressBench, a benchmark of 718 conversations authored by licensed clinicians and crisis counselors, designed to measure the quality of crisis support a model delivers. Each conversation is paired with a clinician-authored rubric of weighted atomic criteria across seven dimensions of crisis support, plus a clinician reference response. A model's score is the share of rubric weight its reply earns. Each criterion names a discrete action a clinician judged necessary, so the score reads directly as the fraction of required care delivered.

Canonical URL: https://benchize.com/benchmarks/scale%3Adistressbench
Last updated: Wed, 21 Oct 2026 18:52:20 GMT

Reproduction and quotation must name Benchize, link to the canonical HTML URL, and retain individual source citations.

Conversational models have become a default place people turn during anxiety, grief, and acute distress. A growing share of that use is emotional, and within it users disclose suicidal ideation and self-harm. When that happens, a model has several jobs at once: recognize the distress, respond with calibrated empathy, de-escalate without endorsing harm, and connect the user to human help and crisis resources. The user may have that conversation only once. We introduceDistressBench, a benchmark of 718 conversations authored by licensed clinicians and crisis counselors, designed to measure the quality of crisis support a model delivers. Each conversation is paired with a clinician-authored rubric of weighted atomic criteria across seven dimensions of crisis support, plus a clinician reference response. A model's score is the share of rubric weight its reply earns. Each criterion names a discrete action a clinician judged necessary, so the score reads directly as the fraction of required care delivered.

Safety & robustness · Models tested: 27 · Score publishers: [scale](https://labs.scale.com/leaderboard/distressbench)

## Model results

| # | Model | Score | ± |  |
| --- | --- | --- | --- | --- |
| 1 | [Muse Spark 1.2](https://benchize.com/models/muse-spark-1-2.md) | 88.39source score | CI 87.46–89.32 |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 88.39 \| source_score \| reported \| Higher is better \| ### Configuration Reasoning effort unknown Model configuration muse-spark-1.2 sourceModelVersion Not reported thinkingEnabled Not reported ### Reported statistics Uncertainty Confidence interval: 87.46–89.32 contaminationWarning Not reported rank 1 reportedConfidenceHalfWidth 0.93 ### Source [Original source](https://labs.scale.com/leaderboard/distressbench) Source scale Original model label muse-spark-1.2 Observed 2026-10-21T16:35:27.403Z Evidence SHA-256 `6ca6a76569e51fde037c8e2610bdf7bc39d99b54c69b6b4dc0dba9ec2a4e24da` Source locator `/rows/0` |  |  |  |  |
| 2 | [Muse Spark 1.3](https://benchize.com/models/muse-spark-1-3.md) | 84.88source score | CI 83.95–85.81 |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 84.88 \| source_score \| reported \| Higher is better \| ### Configuration Reasoning effort unknown Model configuration muse-spark-1.3 sourceModelVersion Not reported thinkingEnabled Not reported ### Reported statistics Uncertainty Confidence interval: 83.95–85.81 contaminationWarning Not reported rank 2 reportedConfidenceHalfWidth 0.93 ### Source [Original source](https://labs.scale.com/leaderboard/distressbench) Source scale Original model label muse-spark-1.3 Observed 2026-10-21T16:35:27.403Z Evidence SHA-256 `6ca6a76569e51fde037c8e2610bdf7bc39d99b54c69b6b4dc0dba9ec2a4e24da` Source locator `/rows/1` |  |  |  |  |
| 3 | [Muse Glimmer 30B](https://benchize.com/models/muse-glimmer-30b.md) | 80.57source score | CI 79.61–81.53 |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 80.57 \| source_score \| reported \| Higher is better \| ### Configuration Reasoning effort unknown Model configuration muse-glimmer-30b sourceModelVersion Not reported thinkingEnabled Not reported ### Reported statistics Uncertainty Confidence interval: 79.61–81.53 contaminationWarning Not reported rank 3 reportedConfidenceHalfWidth 0.96 ### Source [Original source](https://labs.scale.com/leaderboard/distressbench) Source scale Original model label muse-glimmer-30b Observed 2026-10-21T16:35:27.403Z Evidence SHA-256 `6ca6a76569e51fde037c8e2610bdf7bc39d99b54c69b6b4dc0dba9ec2a4e24da` Source locator `/rows/2` |  |  |  |  |
| 4 | [DeepSeek V4.1 Flash](https://benchize.com/models/deepseek-v4-1-flash.md) | 79.78source score | CI 78.77–80.79 |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 79.78 \| source_score \| reported \| Higher is better \| ### Configuration Reasoning effort unknown Model configuration deepseek-v4.1-flash sourceModelVersion deepseek-v4p1-flash thinkingEnabled Not reported ### Reported statistics Uncertainty Confidence interval: 78.77–80.79 contaminationWarning Not reported rank 4 reportedConfidenceHalfWidth 1.01 ### Source [Original source](https://labs.scale.com/leaderboard/distressbench) Source scale Original model label deepseek-v4.1-flash Observed 2026-10-21T16:35:27.403Z Evidence SHA-256 `6ca6a76569e51fde037c8e2610bdf7bc39d99b54c69b6b4dc0dba9ec2a4e24da` Source locator `/rows/3` |  |  |  |  |
| 5 | [Inkling](https://benchize.com/models/inkling.md) | 78.47source score | CI 77.41–79.53 |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 78.47 \| source_score \| reported \| Higher is better \| ### Configuration Reasoning effort unknown Model configuration inkling sourceModelVersion Not reported thinkingEnabled Not reported ### Reported statistics Uncertainty Confidence interval: 77.41–79.53 contaminationWarning Not reported rank 5 reportedConfidenceHalfWidth 1.06 ### Source [Original source](https://labs.scale.com/leaderboard/distressbench) Source scale Original model label inkling Observed 2026-10-21T16:35:27.403Z Evidence SHA-256 `6ca6a76569e51fde037c8e2610bdf7bc39d99b54c69b6b4dc0dba9ec2a4e24da` Source locator `/rows/4` |  |  |  |  |
| 6 | [Qwen3.8 Max](https://benchize.com/models/qwen3-8-max.md) | 77.64source score | CI 76.61–78.67 |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 77.64 \| source_score \| reported \| Higher is better \| ### Configuration Reasoning effort unknown Model configuration qwen3.8-max sourceModelVersion qwen3p8-max thinkingEnabled Not reported ### Reported statistics Uncertainty Confidence interval: 76.61–78.67 contaminationWarning Not reported rank 6 reportedConfidenceHalfWidth 1.03 ### Source [Original source](https://labs.scale.com/leaderboard/distressbench) Source scale Original model label qwen3.8-max Observed 2026-10-21T16:35:27.403Z Evidence SHA-256 `6ca6a76569e51fde037c8e2610bdf7bc39d99b54c69b6b4dc0dba9ec2a4e24da` Source locator `/rows/5` |  |  |  |  |
| 7 | [GPT-5.6 Terra](https://benchize.com/models/gpt-5-6-terra.md) | 76.44source score | CI 75.36–77.52 |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 76.44 \| source_score \| reported \| Higher is better \| ### Configuration Reasoning effort unknown Model configuration gpt-5.6-terra sourceModelVersion gpt-5.6-terra thinkingEnabled Not reported ### Reported statistics Uncertainty Confidence interval: 75.36–77.52 contaminationWarning Not reported rank 8 reportedConfidenceHalfWidth 1.08 ### Source [Original source](https://labs.scale.com/leaderboard/distressbench) Source scale Original model label gpt-5.6-terra Observed 2026-10-21T16:35:27.403Z Evidence SHA-256 `6ca6a76569e51fde037c8e2610bdf7bc39d99b54c69b6b4dc0dba9ec2a4e24da` Source locator `/rows/7` |  |  |  |  |
| 8 | [Claude Opus 5.5](https://benchize.com/models/claude-opus-5-5.md) | 75.58source score | CI 74.38–76.78 |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 75.58 \| source_score \| reported \| Higher is better \| ### Configuration Reasoning effort unknown Model configuration claude-opus-5.5 sourceModelVersion claude-opus-5-5 thinkingEnabled Not reported ### Reported statistics Uncertainty Confidence interval: 74.38–76.78 contaminationWarning Not reported rank 9 reportedConfidenceHalfWidth 1.2 ### Source [Original source](https://labs.scale.com/leaderboard/distressbench) Source scale Original model label claude-opus-5.5 Observed 2026-10-21T16:35:27.403Z Evidence SHA-256 `6ca6a76569e51fde037c8e2610bdf7bc39d99b54c69b6b4dc0dba9ec2a4e24da` Source locator `/rows/8` |  |  |  |  |
| 9 | [GPT-6 Astra](https://benchize.com/models/gpt-6-astra.md) | 74.31source score | CI 73.21–75.41 |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 74.31 \| source_score \| reported \| Higher is better \| ### Configuration Reasoning effort unknown Model configuration gpt-6-astra sourceModelVersion gpt-6-astra thinkingEnabled Not reported ### Reported statistics Uncertainty Confidence interval: 73.21–75.41 contaminationWarning Not reported rank 11 reportedConfidenceHalfWidth 1.1 ### Source [Original source](https://labs.scale.com/leaderboard/distressbench) Source scale Original model label gpt-6-astra Observed 2026-10-21T16:35:27.403Z Evidence SHA-256 `6ca6a76569e51fde037c8e2610bdf7bc39d99b54c69b6b4dc0dba9ec2a4e24da` Source locator `/rows/10` |  |  |  |  |
| 10 | [GPT-5.6 Sol](https://benchize.com/models/gpt-5-6-sol.md) | 73.57source score | CI 72.47–74.67 |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 73.57 \| source_score \| reported \| Higher is better \| ### Configuration Reasoning effort unknown Model configuration gpt-5.6-sol sourceModelVersion gpt-5.6-sol thinkingEnabled Not reported ### Reported statistics Uncertainty Confidence interval: 72.47–74.67 contaminationWarning Not reported rank 12 reportedConfidenceHalfWidth 1.1 ### Source [Original source](https://labs.scale.com/leaderboard/distressbench) Source scale Original model label gpt-5.6-sol Observed 2026-10-21T16:35:27.403Z Evidence SHA-256 `6ca6a76569e51fde037c8e2610bdf7bc39d99b54c69b6b4dc0dba9ec2a4e24da` Source locator `/rows/11` |  |  |  |  |
| 11 | [GPT-5.6 Luna](https://benchize.com/models/gpt-5-6-luna.md) | 72.68source score | CI 71.59–73.77 |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 72.68 \| source_score \| reported \| Higher is better \| ### Configuration Reasoning effort unknown Model configuration gpt-5.6-luna sourceModelVersion gpt-5.6-luna thinkingEnabled Not reported ### Reported statistics Uncertainty Confidence interval: 71.59–73.77 contaminationWarning Not reported rank 15 reportedConfidenceHalfWidth 1.09 ### Source [Original source](https://labs.scale.com/leaderboard/distressbench) Source scale Original model label gpt-5.6-luna Observed 2026-10-21T16:35:27.403Z Evidence SHA-256 `6ca6a76569e51fde037c8e2610bdf7bc39d99b54c69b6b4dc0dba9ec2a4e24da` Source locator `/rows/14` |  |  |  |  |
| 12 | [Claude Fable 5.1](https://benchize.com/models/claude-fable-5-1.md) | 71.55source score | CI 70.23–72.87 |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 71.55 \| source_score \| reported \| Higher is better \| ### Configuration Reasoning effort unknown Model configuration claude-fable-5.1 sourceModelVersion claude-fable-5-1 thinkingEnabled Not reported ### Reported statistics Uncertainty Confidence interval: 70.23–72.87 contaminationWarning Not reported rank 16 reportedConfidenceHalfWidth 1.32 ### Source [Original source](https://labs.scale.com/leaderboard/distressbench) Source scale Original model label claude-fable-5.1 Observed 2026-10-21T16:35:27.403Z Evidence SHA-256 `6ca6a76569e51fde037c8e2610bdf7bc39d99b54c69b6b4dc0dba9ec2a4e24da` Source locator `/rows/15` |  |  |  |  |
| 13 | [Claude Opus 4.8](https://benchize.com/models/claude-opus-4-8.md) | 71.43source score | CI 70.3–72.56 |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 71.43 \| source_score \| reported \| Higher is better \| ### Configuration Reasoning effort unknown Model configuration claude-opus-4.8 sourceModelVersion claude-opus-4-8 thinkingEnabled Not reported ### Reported statistics Uncertainty Confidence interval: 70.3–72.56 contaminationWarning Not reported rank 17 reportedConfidenceHalfWidth 1.13 ### Source [Original source](https://labs.scale.com/leaderboard/distressbench) Source scale Original model label claude-opus-4.8 Observed 2026-10-21T16:35:27.403Z Evidence SHA-256 `6ca6a76569e51fde037c8e2610bdf7bc39d99b54c69b6b4dc0dba9ec2a4e24da` Source locator `/rows/16` |  |  |  |  |
| 14 | [Gemini 3.8 Flash](https://benchize.com/models/gemini-3-8-flash.md) | 70.74source score | CI 69.51–71.97 |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 70.74 \| source_score \| reported \| Higher is better \| ### Configuration Reasoning effort unknown Model configuration gemini-3.8-flash sourceModelVersion gemini-3.8-flash thinkingEnabled Not reported ### Reported statistics Uncertainty Confidence interval: 69.51–71.97 contaminationWarning Not reported rank 19 reportedConfidenceHalfWidth 1.23 ### Source [Original source](https://labs.scale.com/leaderboard/distressbench) Source scale Original model label gemini-3.8-flash Observed 2026-10-21T16:35:27.403Z Evidence SHA-256 `6ca6a76569e51fde037c8e2610bdf7bc39d99b54c69b6b4dc0dba9ec2a4e24da` Source locator `/rows/18` |  |  |  |  |
| 15 | [DeepSeek V4 Flash](https://benchize.com/models/deepseek-v4-flash.md) | 69.66source score | CI 68.44–70.88 |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 69.66 \| source_score \| reported \| Higher is better \| ### Configuration Reasoning effort unknown Model configuration deepseek-v4-flash sourceModelVersion deepseek-v4-flash thinkingEnabled Not reported ### Reported statistics Uncertainty Confidence interval: 68.44–70.88 contaminationWarning Not reported rank 20 reportedConfidenceHalfWidth 1.22 ### Source [Original source](https://labs.scale.com/leaderboard/distressbench) Source scale Original model label deepseek-v4-flash Observed 2026-10-21T16:35:27.403Z Evidence SHA-256 `6ca6a76569e51fde037c8e2610bdf7bc39d99b54c69b6b4dc0dba9ec2a4e24da` Source locator `/rows/19` |  |  |  |  |
| 16 | [Claude Fable 5](https://benchize.com/models/claude-fable-5.md) | 69.43source score | CI 67.68–71.18 |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 69.43 \| source_score \| reported \| Higher is better \| ### Configuration Reasoning effort unknown Model configuration claude-fable-5 sourceModelVersion claude-fable-5 thinkingEnabled Not reported ### Reported statistics Uncertainty Confidence interval: 67.68–71.18 contaminationWarning Not reported rank 21 reportedConfidenceHalfWidth 1.75 ### Source [Original source](https://labs.scale.com/leaderboard/distressbench) Source scale Original model label claude-fable-5 Observed 2026-10-21T16:35:27.403Z Evidence SHA-256 `6ca6a76569e51fde037c8e2610bdf7bc39d99b54c69b6b4dc0dba9ec2a4e24da` Source locator `/rows/20` |  |  |  |  |
| 17 | [DeepSeek V4 Pro](https://benchize.com/models/deepseek-v4-pro.md) | 69.19source score | CI 67.95–70.43 |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 69.19 \| source_score \| reported \| Higher is better \| ### Configuration Reasoning effort unknown Model configuration deepseek-v4-pro sourceModelVersion deepseek-v4-pro thinkingEnabled Not reported ### Reported statistics Uncertainty Confidence interval: 67.95–70.43 contaminationWarning Not reported rank 22 reportedConfidenceHalfWidth 1.24 ### Source [Original source](https://labs.scale.com/leaderboard/distressbench) Source scale Original model label deepseek-v4-pro Observed 2026-10-21T16:35:27.403Z Evidence SHA-256 `6ca6a76569e51fde037c8e2610bdf7bc39d99b54c69b6b4dc0dba9ec2a4e24da` Source locator `/rows/21` |  |  |  |  |
| 18 | [Gemini 3.7 Flash](https://benchize.com/models/gemini-3-7-flash.md) | 68.95source score | CI 67.67–70.23 |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 68.95 \| source_score \| reported \| Higher is better \| ### Configuration Reasoning effort unknown Model configuration gemini-3.7-flash sourceModelVersion gemini-3.7-flash thinkingEnabled Not reported ### Reported statistics Uncertainty Confidence interval: 67.67–70.23 contaminationWarning Not reported rank 23 reportedConfidenceHalfWidth 1.28 ### Source [Original source](https://labs.scale.com/leaderboard/distressbench) Source scale Original model label gemini-3.7-flash Observed 2026-10-21T16:35:27.403Z Evidence SHA-256 `6ca6a76569e51fde037c8e2610bdf7bc39d99b54c69b6b4dc0dba9ec2a4e24da` Source locator `/rows/22` |  |  |  |  |
| 19 | [Gemini 3.1 Pro](https://benchize.com/models/gemini-3-1-pro.md) | 68.76source score | CI 67.46–70.06 |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 68.76 \| source_score \| reported \| Higher is better \| ### Configuration Reasoning effort unknown Model configuration gemini-3.1-pro sourceModelVersion gemini-3.1-pro-preview thinkingEnabled Not reported ### Reported statistics Uncertainty Confidence interval: 67.46–70.06 contaminationWarning Not reported rank 24 reportedConfidenceHalfWidth 1.3 ### Source [Original source](https://labs.scale.com/leaderboard/distressbench) Source scale Original model label gemini-3.1-pro Observed 2026-10-21T16:35:27.403Z Evidence SHA-256 `6ca6a76569e51fde037c8e2610bdf7bc39d99b54c69b6b4dc0dba9ec2a4e24da` Source locator `/rows/23` |  |  |  |  |
| 20 | [Claude Sonnet 5](https://benchize.com/models/claude-sonnet-5.md) | 67.98source score | CI 66.82–69.14 |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 67.98 \| source_score \| reported \| Higher is better \| ### Configuration Reasoning effort unknown Model configuration claude-sonnet-5 sourceModelVersion claude-sonnet-5 thinkingEnabled Not reported ### Reported statistics Uncertainty Confidence interval: 66.82–69.14 contaminationWarning Not reported rank 25 reportedConfidenceHalfWidth 1.16 ### Source [Original source](https://labs.scale.com/leaderboard/distressbench) Source scale Original model label claude-sonnet-5 Observed 2026-10-21T16:35:27.403Z Evidence SHA-256 `6ca6a76569e51fde037c8e2610bdf7bc39d99b54c69b6b4dc0dba9ec2a4e24da` Source locator `/rows/24` |  |  |  |  |
| 21 | [GPT-6 Sol](https://benchize.com/models/gpt-6-sol.md) | 67.96source score | CI 66.79–69.13 |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 67.96 \| source_score \| reported \| Higher is better \| ### Configuration Reasoning effort unknown Model configuration gpt-6-sol sourceModelVersion gpt-6-sol thinkingEnabled Not reported ### Reported statistics Uncertainty Confidence interval: 66.79–69.13 contaminationWarning Not reported rank 26 reportedConfidenceHalfWidth 1.17 ### Source [Original source](https://labs.scale.com/leaderboard/distressbench) Source scale Original model label gpt-6-sol Observed 2026-10-21T16:35:27.403Z Evidence SHA-256 `6ca6a76569e51fde037c8e2610bdf7bc39d99b54c69b6b4dc0dba9ec2a4e24da` Source locator `/rows/25` |  |  |  |  |
| 22 | [Gemini 3.5 Flash](https://benchize.com/models/gemini-3-5-flash.md) | 66.89source score | CI 65.61–68.17 |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 66.89 \| source_score \| reported \| Higher is better \| ### Configuration Reasoning effort unknown Model configuration gemini-3.5-flash sourceModelVersion gemini-3.5-flash thinkingEnabled Not reported ### Reported statistics Uncertainty Confidence interval: 65.61–68.17 contaminationWarning Not reported rank 27 reportedConfidenceHalfWidth 1.28 ### Source [Original source](https://labs.scale.com/leaderboard/distressbench) Source scale Original model label gemini-3.5-flash Observed 2026-10-21T16:35:27.403Z Evidence SHA-256 `6ca6a76569e51fde037c8e2610bdf7bc39d99b54c69b6b4dc0dba9ec2a4e24da` Source locator `/rows/26` |  |  |  |  |
| 23 | [GPT-6 Luna](https://benchize.com/models/gpt-6-luna.md) | 66.66source score | CI 65.52–67.8 |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 66.66 \| source_score \| reported \| Higher is better \| ### Configuration Reasoning effort unknown Model configuration gpt-6-luna sourceModelVersion gpt-6-luna thinkingEnabled Not reported ### Reported statistics Uncertainty Confidence interval: 65.52–67.8 contaminationWarning Not reported rank 28 reportedConfidenceHalfWidth 1.14 ### Source [Original source](https://labs.scale.com/leaderboard/distressbench) Source scale Original model label gpt-6-luna Observed 2026-10-21T16:35:27.403Z Evidence SHA-256 `6ca6a76569e51fde037c8e2610bdf7bc39d99b54c69b6b4dc0dba9ec2a4e24da` Source locator `/rows/27` |  |  |  |  |
| 24 | [gpt-oss-120b](https://benchize.com/models/gpt-oss-120b.md) | 63.68source score | CI 62.18–65.18 |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 63.68 \| source_score \| reported \| Higher is better \| ### Configuration Reasoning effort unknown Model configuration gpt-oss-120b sourceModelVersion gpt-oss-120b thinkingEnabled Not reported ### Reported statistics Uncertainty Confidence interval: 62.18–65.18 contaminationWarning Not reported rank 29 reportedConfidenceHalfWidth 1.5 ### Source [Original source](https://labs.scale.com/leaderboard/distressbench) Source scale Original model label gpt-oss-120b Observed 2026-10-21T16:35:27.403Z Evidence SHA-256 `6ca6a76569e51fde037c8e2610bdf7bc39d99b54c69b6b4dc0dba9ec2a4e24da` Source locator `/rows/28` |  |  |  |  |
| 25 | [Mistral Large](https://benchize.com/models/mistral-large.md) | 59.08source score | CI 57.66–60.5 |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 59.08 \| source_score \| reported \| Higher is better \| ### Configuration Reasoning effort unknown Model configuration mistral-large sourceModelVersion mistral-large-2512 thinkingEnabled Not reported ### Reported statistics Uncertainty Confidence interval: 57.66–60.5 contaminationWarning Not reported rank 30 reportedConfidenceHalfWidth 1.42 ### Source [Original source](https://labs.scale.com/leaderboard/distressbench) Source scale Original model label mistral-large Observed 2026-10-21T16:35:27.403Z Evidence SHA-256 `6ca6a76569e51fde037c8e2610bdf7bc39d99b54c69b6b4dc0dba9ec2a4e24da` Source locator `/rows/29` |  |  |  |  |
| 26 | [Mistral Large 3](https://benchize.com/models/mistral-large-3.md) | 58.77source score | CI 57.4–60.14 |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 58.77 \| source_score \| reported \| Higher is better \| ### Configuration Reasoning effort unknown Model configuration mistral-large-3 sourceModelVersion mistral-large-3-675b-instruct thinkingEnabled Not reported ### Reported statistics Uncertainty Confidence interval: 57.4–60.14 contaminationWarning Not reported rank 31 reportedConfidenceHalfWidth 1.37 ### Source [Original source](https://labs.scale.com/leaderboard/distressbench) Source scale Original model label mistral-large-3 Observed 2026-10-21T16:35:27.403Z Evidence SHA-256 `6ca6a76569e51fde037c8e2610bdf7bc39d99b54c69b6b4dc0dba9ec2a4e24da` Source locator `/rows/30` |  |  |  |  |
| 27 | [Llama 4 Maverick](https://benchize.com/models/llama-4-maverick.md) | 41.25source score | CI 40.05–42.45 |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 41.25 \| source_score \| reported \| Higher is better \| ### Configuration Reasoning effort unknown Model configuration llama-4-maverick sourceModelVersion llama4-maverick-17b-instruct-v1:0 thinkingEnabled Not reported ### Reported statistics Uncertainty Confidence interval: 40.05–42.45 contaminationWarning Not reported rank 34 reportedConfidenceHalfWidth 1.2 ### Source [Original source](https://labs.scale.com/leaderboard/distressbench) Source scale Original model label llama-4-maverick Observed 2026-10-21T16:35:27.403Z Evidence SHA-256 `6ca6a76569e51fde037c8e2610bdf7bc39d99b54c69b6b4dc0dba9ec2a4e24da` Source locator `/rows/32` |  |  |  |  |

---
Cite as: DistressBench — Benchize, https://benchize.com/benchmarks/scale%3Adistressbench (retrieved 2026-10-21)