PAIRS (Parallel Images)
Models answer questions about people in paired images. The evaluation tests whether changing demographic characteristics changes judgments about the subjects.
- Made by
- Kathleen C. Fraser and colleagues
- Released
- February 2024
- Kind
- Bias, fairness and politics
- Level
- Everyday
- Scored
- Checked automatically
- Answers
- Multiple choice
- Tasks
- All public
- Measures
- Propensity
- Behavior
- Bias
- Language
- English
- Interaction
- Single turn
- Refresh
- One fixed set
- Run by
- Anyone
- In views
- Vision
- Strongest model tested
- o3 #84 on the capability index
- Last model added
- April 2025
- Scores refreshed
- November 20th, 2026
GPT-4o leads with 97.2% among models released by April 2025 GPT-4o leads with 97.2% among models released by April 2025
Best score per entry from VHELM’s leaderboard, read on November 20th, 2026. VHELM reports the percentage of responses that match the prefix of a correct reference after light processing. Where a model ran with several settings or harnesses, we keep its best. 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 April 2025, 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
- 97.2 GPT-4o
- 95.4 GPT-4o
- 93.0 GPT-4.1 nano
- 92.3
Gemini 1.5 Pro
- 92.2 o1
- 91.6 GPT-4 Turbo
- 91.4
Gemini 1.5 Pro
- 89.8 GPT-4.5
- 89.2 GPT-4.1 mini
- 87.3 GPT-4o
- 87.2
Gemini 2.0 Pro
- 86.1 GPT-4 Turbo
- 84.8
Gemini 2.0 Flash-Lite
- 83.8
Gemini 2.5 Pro
- 82.8 Qwen2-VL-72B-Instruct
- 80.1
Qwen2.5-VL-32B
- 79.6 GPT-4o mini
- 78.8
Gemini 1.0 Pro Vision
- 77.4 GPT-4.1
- 76.5 o3
- 75.6
Qwen2.5-VL-72B
- 74.0 palmyra-vision-003
- 74.0
Gemini 1.5 Flash
- 74.0
Gemini 1.5 Flash
- 73.4 o4-mini
- 71.5
Gemini 2.0 Flash
- 68.2
Llama 4 Maverick
- 65.4 Claude 3.5 Sonnet (October 2024)
- 62.3
Qwen2.5-VL-7B
- 61.4 Claude 3.5 Sonnet
- 58.7 Claude 3 Opus
- 55.4
Llama 4 Scout
- 48.8 Qwen2-VL-7B-Instruct
- 48.7
Idefics2-8B
- 46.6
Pixtral 12B
- 36.8
Qwen2.5-VL-3B
- 14.0
PaliGemma-3B-mix-448
- 13.0 Claude 3 Sonnet
- 8.03 Claude 3 Haiku
- 8.00
IDEFICS-9B-Instruct
- 7.33
IDEFICS-80B-Instruct
- 0.0 paligemma-3b-mix-224
- 0.0 Qwen-VL-Chat
- 1 GPT-4o Nov 2024 97.2%
- 2 GPT-4o Aug 2024 95.4%
- 3 GPT-4.1 nano 93%
- 4
Gemini 1.5 Pro Sep 2024 92.3%
- 5 o1 92.2%
- 6 GPT-4 Turbo Nov 2023 91.6%
- 7
Gemini 1.5 Pro May 2024, BLOCK_NONE safety 91.4%
- 8 GPT-4.5 89.8%
- 9 GPT-4.1 mini 89.2%
- 10 GPT-4o May 2024 87.3%
- 11
Gemini 2.0 Pro 87.2%
- 12 GPT-4 Turbo Apr 2024 86.1%
- 13
Gemini 2.0 Flash-Lite 84.8%
- 14
Gemini 2.5 Pro Mar 2025 83.8%
- 15 Qwen2-VL-72B-Instruct 82.8%
- 16
Qwen2.5-VL-32B 80.1%
- 17 GPT-4o mini 79.6%
- 18
Gemini 1.0 Pro Vision 78.8%
- 19 GPT-4.1 77.4%
- 20 o3 High reasoning effort 76.5%
- 21
Qwen2.5-VL-72B 75.6%
- 22 palmyra-vision-003 74%
- 23
Gemini 1.5 Flash May 2024, BLOCK_NONE safety 74%
- 24
Gemini 1.5 Flash Sep 2024 74%
- 25 o4-mini High reasoning effort 73.4%
- 26
Gemini 2.0 Flash Feb 2025 71.5%
- 27
Llama 4 Maverick 68.2%
- 28 Claude 3.5 Sonnet (October 2024) 65.4%
- 29
Qwen2.5-VL-7B 62.3%
- 30 Claude 3.5 Sonnet 61.4%
- 31 Claude 3 Opus 58.7%
- 32
Llama 4 Scout 55.4%
- 33 Qwen2-VL-7B-Instruct 48.8%
- 34
Idefics2-8B 48.7%
- 35
Pixtral 12B 46.6%
- 36
Qwen2.5-VL-3B 36.8%
- 37
PaliGemma-3B-mix-448 14%
- 38 Claude 3 Sonnet 13%
- 39 Claude 3 Haiku 8.03%
- 40
IDEFICS-9B-Instruct 8.00%
- 41
IDEFICS-80B-Instruct 7.33%
- 42 paligemma-3b-mix-224 0%
- 43 Qwen-VL-Chat 0%
Source: VHELM’s leaderboard
- 97.2 GPT-4o
- 95.4 GPT-4o
- 93.0 GPT-4.1 nano
- 92.3
Gemini 1.5 Pro
- 92.2 o1
- 91.6 GPT-4 Turbo
- 91.4
Gemini 1.5 Pro
- 89.8 GPT-4.5
- 89.2 GPT-4.1 mini
- 87.3 GPT-4o
- 87.2
Gemini 2.0 Pro
- 86.1 GPT-4 Turbo
- 84.8
Gemini 2.0 Flash-Lite
- 83.8
Gemini 2.5 Pro
- 82.8 Qwen2-VL-72B-Instruct
- 80.1
Qwen2.5-VL-32B
- 79.6 GPT-4o mini
- 78.8
Gemini 1.0 Pro Vision
- 77.4 GPT-4.1
- 76.5 o3
- 75.6
Qwen2.5-VL-72B
- 74.0
Gemini 1.5 Flash
- 74.0
Gemini 1.5 Flash
- 73.4 o4-mini
- 71.5
Gemini 2.0 Flash
- 68.2
Llama 4 Maverick
- 65.4 Claude 3.5 Sonnet (October 2024)
- 62.3
Qwen2.5-VL-7B
- 61.4 Claude 3.5 Sonnet
- 58.7 Claude 3 Opus
- 55.4
Llama 4 Scout
- 48.8 Qwen2-VL-7B-Instruct
- 48.7
Idefics2-8B
- 46.6
Pixtral 12B
- 36.8
Qwen2.5-VL-3B
- 14.0
PaliGemma-3B-mix-448
- 13.0 Claude 3 Sonnet
- 8.03 Claude 3 Haiku
- 8.00
IDEFICS-9B-Instruct
- 7.33
IDEFICS-80B-Instruct
- 0.0 Qwen-VL-Chat
- 1 GPT-4o Nov 2024 97.2%
- 2 GPT-4o Aug 2024 95.4%
- 3 GPT-4.1 nano 93%
- 4
Gemini 1.5 Pro Sep 2024 92.3%
- 5 o1 92.2%
- 6 GPT-4 Turbo Nov 2023 91.6%
- 7
Gemini 1.5 Pro May 2024, BLOCK_NONE safety 91.4%
- 8 GPT-4.5 89.8%
- 9 GPT-4.1 mini 89.2%
- 10 GPT-4o May 2024 87.3%
- 11
Gemini 2.0 Pro 87.2%
- 12 GPT-4 Turbo Apr 2024 86.1%
- 13
Gemini 2.0 Flash-Lite 84.8%
- 14
Gemini 2.5 Pro Mar 2025 83.8%
- 15 Qwen2-VL-72B-Instruct 82.8%
- 16
Qwen2.5-VL-32B 80.1%
- 17 GPT-4o mini 79.6%
- 18
Gemini 1.0 Pro Vision 78.8%
- 19 GPT-4.1 77.4%
- 20 o3 High reasoning effort 76.5%
- 21
Qwen2.5-VL-72B 75.6%
- 22
Gemini 1.5 Flash May 2024, BLOCK_NONE safety 74%
- 23
Gemini 1.5 Flash Sep 2024 74%
- 24 o4-mini High reasoning effort 73.4%
- 25
Gemini 2.0 Flash Feb 2025 71.5%
- 26
Llama 4 Maverick 68.2%
- 27 Claude 3.5 Sonnet (October 2024) 65.4%
- 28
Qwen2.5-VL-7B 62.3%
- 29 Claude 3.5 Sonnet 61.4%
- 30 Claude 3 Opus 58.7%
- 31
Llama 4 Scout 55.4%
- 32 Qwen2-VL-7B-Instruct 48.8%
- 33
Idefics2-8B 48.7%
- 34
Pixtral 12B 46.6%
- 35
Qwen2.5-VL-3B 36.8%
- 36
PaliGemma-3B-mix-448 14%
- 37 Claude 3 Sonnet 13%
- 38 Claude 3 Haiku 8.03%
- 39
IDEFICS-9B-Instruct 8.00%
- 40
IDEFICS-80B-Instruct 7.33%
- 41 Qwen-VL-Chat 0%
Source: VHELM’s leaderboard
| # | Entry | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | GPT-4o Nov 2024 | 97.2% | Nov 2024 | Official leaderboard | |
| 2 | GPT-4o Aug 2024 | 95.4% | Aug 2024 | Official leaderboard | |
| 3 | GPT-4.1 nano | 93% | Apr 2025 | Official leaderboard | |
| 4 | | 92.3% | Sep 2024 | Official leaderboard | |
| 5 | o1 | 92.2% | Dec 2024 | Official leaderboard | |
| 6 | GPT-4 Turbo Nov 2023 | 91.6% | Nov 2023 | Official leaderboard | |
| 7 | | 91.4% | BLOCK_NONE safety | May 2024 | Official leaderboard |
| 8 | GPT-4.5 | 89.8% | Feb 2025 | Official leaderboard | |
| 9 | GPT-4.1 mini | 89.2% | Apr 2025 | Official leaderboard | |
| 10 | GPT-4o May 2024 | 87.3% | May 2024 | Official leaderboard | |
| 11 | | 87.2% | Feb 2025 | Official leaderboard | |
| 12 | GPT-4 Turbo Apr 2024 | 86.1% | Apr 2024 | Official leaderboard | |
| 13 | | 84.8% | Feb 2025 | Official leaderboard | |
| 14 | | 83.8% | Mar 2025 | Official leaderboard | |
| 15 | Qwen2-VL-72B-Instruct | 82.8% | Aug 2024 | Official leaderboard | |
| 16 | | 80.1% | Mar 2025 | Official leaderboard | |
| 17 | GPT-4o mini | 79.6% | Jul 2024 | Official leaderboard | |
| 18 | | 78.8% | Jan 2024 | Official leaderboard | |
| 19 | GPT-4.1 | 77.4% | Apr 2025 | Official leaderboard | |
| 20 | o3 | 76.5% | High reasoning effort | Apr 2025 | Official leaderboard |
| 21 | | 75.6% | Jan 2025 | Official leaderboard | |
| 22 | palmyra-vision-003 | 74% | Official leaderboard | ||
| 23 | | 74% | BLOCK_NONE safety | May 2024 | Official leaderboard |
| 24 | | 74% | Aug 2024 | Official leaderboard | |
| 25 | o4-mini | 73.4% | High reasoning effort | Apr 2025 | Official leaderboard |
| 26 | | 71.5% | Feb 2025 | Official leaderboard | |
| 27 | | 68.2% | Apr 2025 | Official leaderboard | |
| 28 | Claude 3.5 Sonnet (October 2024) | 65.4% | Oct 2024 | Official leaderboard | |
| 29 | | 62.3% | Jan 2025 | Official leaderboard | |
| 30 | Claude 3.5 Sonnet | 61.4% | Jun 2024 | Official leaderboard | |
| 31 | Claude 3 Opus | 58.7% | Feb 2024 | Official leaderboard | |
| 32 | | 55.4% | Apr 2025 | Official leaderboard | |
| 33 | Qwen2-VL-7B-Instruct | 48.8% | Aug 2024 | Official leaderboard | |
| 34 | | 48.7% | Apr 2024 | Official leaderboard | |
| 35 | | 46.6% | Sep 2024 | Official leaderboard | |
| 36 | | 36.8% | Jan 2025 | Official leaderboard | |
| 37 | | 14% | May 2024 | Official leaderboard | |
| 38 | Claude 3 Sonnet | 13% | Feb 2024 | Official leaderboard | |
| 39 | Claude 3 Haiku | 8.03% | Mar 2024 | Official leaderboard | |
| 40 | | 8.00% | Jul 2023 | Official leaderboard | |
| 41 | | 7.33% | Jul 2023 | Official leaderboard | |
| 42 | paligemma-3b-mix-224 | 0% | Official leaderboard | ||
| 43 | Qwen-VL-Chat | 0% | Aug 2023 | Official leaderboard |
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | GPT-4o Nov 2024 | 97.2% | Nov 2024 | Official leaderboard | |
| 2 | GPT-4o Aug 2024 | 95.4% | Aug 2024 | Official leaderboard | |
| 3 | GPT-4.1 nano | 93% | Apr 2025 | Official leaderboard | |
| 4 | | 92.3% | Sep 2024 | Official leaderboard | |
| 5 | o1 | 92.2% | Dec 2024 | Official leaderboard | |
| 6 | GPT-4 Turbo Nov 2023 | 91.6% | Nov 2023 | Official leaderboard | |
| 7 | | 91.4% | BLOCK_NONE safety | May 2024 | Official leaderboard |
| 8 | GPT-4.5 | 89.8% | Feb 2025 | Official leaderboard | |
| 9 | GPT-4.1 mini | 89.2% | Apr 2025 | Official leaderboard | |
| 10 | GPT-4o May 2024 | 87.3% | May 2024 | Official leaderboard | |
| 11 | | 87.2% | Feb 2025 | Official leaderboard | |
| 12 | GPT-4 Turbo Apr 2024 | 86.1% | Apr 2024 | Official leaderboard | |
| 13 | | 84.8% | Feb 2025 | Official leaderboard | |
| 14 | | 83.8% | Mar 2025 | Official leaderboard | |
| 15 | Qwen2-VL-72B-Instruct | 82.8% | Aug 2024 | Official leaderboard | |
| 16 | | 80.1% | Mar 2025 | Official leaderboard | |
| 17 | GPT-4o mini | 79.6% | Jul 2024 | Official leaderboard | |
| 18 | | 78.8% | Jan 2024 | Official leaderboard | |
| 19 | GPT-4.1 | 77.4% | Apr 2025 | Official leaderboard | |
| 20 | o3 | 76.5% | High reasoning effort | Apr 2025 | Official leaderboard |
| 21 | | 75.6% | Jan 2025 | Official leaderboard | |
| 22 | | 74% | BLOCK_NONE safety | May 2024 | Official leaderboard |
| 23 | | 74% | Aug 2024 | Official leaderboard | |
| 24 | o4-mini | 73.4% | High reasoning effort | Apr 2025 | Official leaderboard |
| 25 | | 71.5% | Feb 2025 | Official leaderboard | |
| 26 | | 68.2% | Apr 2025 | Official leaderboard | |
| 27 | Claude 3.5 Sonnet (October 2024) | 65.4% | Oct 2024 | Official leaderboard | |
| 28 | | 62.3% | Jan 2025 | Official leaderboard | |
| 29 | Claude 3.5 Sonnet | 61.4% | Jun 2024 | Official leaderboard | |
| 30 | Claude 3 Opus | 58.7% | Feb 2024 | Official leaderboard | |
| 31 | | 55.4% | Apr 2025 | Official leaderboard | |
| 32 | Qwen2-VL-7B-Instruct | 48.8% | Aug 2024 | Official leaderboard | |
| 33 | | 48.7% | Apr 2024 | Official leaderboard | |
| 34 | | 46.6% | Sep 2024 | Official leaderboard | |
| 35 | | 36.8% | Jan 2025 | Official leaderboard | |
| 36 | | 14% | May 2024 | Official leaderboard | |
| 37 | Claude 3 Sonnet | 13% | Feb 2024 | Official leaderboard | |
| 38 | Claude 3 Haiku | 8.03% | Mar 2024 | Official leaderboard | |
| 39 | | 8.00% | Jul 2023 | Official leaderboard | |
| 40 | | 7.33% | Jul 2023 | Official leaderboard | |
| 41 | Qwen-VL-Chat | 0% | Aug 2023 | Official leaderboard |
3 models have held number one on VHELM’s leaderboard since its release in February 2024
Lab claims count from the day the card came out, independent runs from the model’s release.
Source: VHELM’s leaderboard
GPT-4 Turbo on the official leaderboard passed 85% in November 2023
Models released before it came out already scored that high. All the scores we have, one dot each, so a model tested at several settings appears more than once.
Source: VHELM’s 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 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%
- Covert misbehavior (Thinking Machines) In use Agentic tasks that check whether a model misbehaves and conceals it. Scored by the rate of covert misbehavior.
Inkling 2 Mini 8.30%
- Destructive actions (Thinking Machines) In use Agentic tasks that check whether a model takes destructive actions it was not authorized to take. Scored by the rate of destructive actions.
Inkling 2 Small 4.70%
- Misbehavior (Thinking Machines) In use Agentic tasks that check whether a model fabricates results, ignores instructions or hides what it did when a task is hard. Scored by the rate of misbehavior.
Inkling 2 Mini 9.80%
Data audit
All 8 checks pass. Every benchmark
Catalog record 2 pass
- Pass Task access All tasks are public.
- Pass Confidence Checked against a primary source.
Scores 2 pass 3 notes
- Pass Independent runs Scores from the official leaderboard (43 entries).
- Pass Models scored 41 models we track, plus 2 other entries on its leaderboard.
- Note Newest score The newest came 19 months ago, in April 2025, for o3. That’s expected for a saturated test.
- Note Frontier models None of the top 10 has a score, which is fine for a retired test. The strongest tested is o3 (#84).
- Note Headroom Saturated. The best is 97.2%, past 85% since November 2023.
Leaderboard import 3 pass
- Pass Importer We read the board on VHELM: 46 rows for models we track and 2 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.