provider
stringclasses 54
values | name
stringclasses 186
values | size
stringclasses 120
values | variant
stringclasses 110
values | version
stringclasses 110
values | sector
stringclasses 4
values | openness
stringclasses 2
values | region
stringclasses 5
values | country
stringclasses 13
values | source_id
stringclasses 434
values | is_first_party
bool 2
classes | category
int64 1
7
| year
int64 2.02k
2.03k
| metadata
stringclasses 433
values | score
float64 0
3
| is_model_release
bool 2
classes |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
Meta
|
llama-2
|
7B
|
chat
| null |
Industry
|
open
|
North America
|
United States
|
first-party-or-cooperative-evals_47
| false
| 1
| 2,025
|
{'title': 'Raising the Bar: Investigating the Values of Large Language Models via Generative Evolving Testing', 'url': 'https://arxiv.org/pdf/2406.14230'}
| 3
| false
|
Google
|
gemini-1.0
| null |
pro
| null |
Industry
|
closed
|
North America
|
United States
|
first-party-or-cooperative-evals_47
| false
| 1
| 2,025
|
{'title': 'Raising the Bar: Investigating the Values of Large Language Models via Generative Evolving Testing', 'url': 'https://arxiv.org/pdf/2406.14230'}
| 3
| false
|
Meta
|
llama-3.1
|
8B
|
instruct
| null |
Industry
|
open
|
North America
|
United States
|
first-party-or-cooperative-evals_48
| false
| 1
| 2,025
|
{'title': 'Controllable Safety Alignment: Inference-Time Adaptation to Diverse Safety Requirements ', 'url': 'https://arxiv.org/pdf/2410.08968'}
| 3
| false
|
OpenAI
|
gpt-4
| null |
mini
| null |
Industry
|
closed
|
North America
|
United States
|
first-party-or-cooperative-evals_48
| false
| 1
| 2,025
|
{'title': 'Controllable Safety Alignment: Inference-Time Adaptation to Diverse Safety Requirements ', 'url': 'https://arxiv.org/pdf/2410.08968'}
| 3
| false
|
OpenAI
|
gpt-4
| null | null | null |
Industry
|
closed
|
North America
|
United States
|
first-party-or-cooperative-evals_48
| false
| 1
| 2,025
|
{'title': 'Controllable Safety Alignment: Inference-Time Adaptation to Diverse Safety Requirements ', 'url': 'https://arxiv.org/pdf/2410.08968'}
| 3
| false
|
Meta
|
llama-3.1
|
8B
|
instruct
| null |
Industry
|
open
|
North America
|
United States
|
first-party-or-cooperative-evals_48
| false
| 2
| 2,025
|
{'title': 'Controllable Safety Alignment: Inference-Time Adaptation to Diverse Safety Requirements ', 'url': 'https://arxiv.org/pdf/2410.08968'}
| 3
| false
|
OpenAI
|
gpt-4
| null | null | null |
Industry
|
closed
|
North America
|
United States
|
first-party-or-cooperative-evals_48
| false
| 2
| 2,025
|
{'title': 'Controllable Safety Alignment: Inference-Time Adaptation to Diverse Safety Requirements ', 'url': 'https://arxiv.org/pdf/2410.08968'}
| 3
| false
|
OpenAI
|
gpt-4
| null |
mini
| null |
Industry
|
closed
|
North America
|
United States
|
first-party-or-cooperative-evals_48
| false
| 2
| 2,025
|
{'title': 'Controllable Safety Alignment: Inference-Time Adaptation to Diverse Safety Requirements ', 'url': 'https://arxiv.org/pdf/2410.08968'}
| 3
| false
|
Microsoft
|
phi-3
| null | null | null |
Industry
|
open
|
North America
|
United States
|
first-party-or-cooperative-evals_49
| true
| 2
| 2,025
|
{'title': 'Lessons From Red Teaming 100 Generative AI Products', 'url': 'https://ashy-coast-00aeb501e.6.azurestaticapps.net/MS_AIRT_Lessons_eBook.pdf', 'release_date': '2025-01-13'}
| 1
| false
|
OpenAI
|
gpt-4
| null | null | null |
Industry
|
closed
|
North America
|
United States
|
first-party-or-cooperative-evals_49
| false
| 2
| 2,025
|
{'title': 'Lessons From Red Teaming 100 Generative AI Products', 'url': 'https://ashy-coast-00aeb501e.6.azurestaticapps.net/MS_AIRT_Lessons_eBook.pdf', 'release_date': '2025-01-13'}
| 1
| false
|
Microsoft
|
phi-3.5
| null | null | null |
Industry
|
open
|
North America
|
United States
|
first-party-or-cooperative-evals_49
| true
| 2
| 2,025
|
{'title': 'Lessons From Red Teaming 100 Generative AI Products', 'url': 'https://ashy-coast-00aeb501e.6.azurestaticapps.net/MS_AIRT_Lessons_eBook.pdf', 'release_date': '2025-01-13'}
| 1
| false
|
Microsoft
|
phi-3
| null | null | null |
Industry
|
open
|
North America
|
United States
|
first-party-or-cooperative-evals_49
| true
| 1
| 2,025
|
{'title': 'Lessons From Red Teaming 100 Generative AI Products', 'url': 'https://ashy-coast-00aeb501e.6.azurestaticapps.net/MS_AIRT_Lessons_eBook.pdf', 'release_date': '2025-01-13'}
| 1
| false
|
OpenAI
|
gpt-4
| null | null | null |
Industry
|
closed
|
North America
|
United States
|
first-party-or-cooperative-evals_49
| false
| 1
| 2,025
|
{'title': 'Lessons From Red Teaming 100 Generative AI Products', 'url': 'https://ashy-coast-00aeb501e.6.azurestaticapps.net/MS_AIRT_Lessons_eBook.pdf', 'release_date': '2025-01-13'}
| 1
| false
|
Microsoft
|
phi-3.5
| null | null | null |
Industry
|
open
|
North America
|
United States
|
first-party-or-cooperative-evals_49
| true
| 1
| 2,025
|
{'title': 'Lessons From Red Teaming 100 Generative AI Products', 'url': 'https://ashy-coast-00aeb501e.6.azurestaticapps.net/MS_AIRT_Lessons_eBook.pdf', 'release_date': '2025-01-13'}
| 1
| false
|
OpenAI
|
o1
| null |
preview
| null |
Industry
|
closed
|
North America
|
United States
|
first-party-or-cooperative-evals_50
| true
| 3
| 2,024
|
{'title': 'Evaluating fairness in ChatGPT', 'url': 'https://cdn.openai.com/papers/first-person-fairness-in-chatbots.pdf', 'release_date': '2024-10-15'}
| 3
| false
|
OpenAI
|
gpt-4
| null | null | null |
Industry
|
closed
|
North America
|
United States
|
first-party-or-cooperative-evals_50
| true
| 3
| 2,024
|
{'title': 'Evaluating fairness in ChatGPT', 'url': 'https://cdn.openai.com/papers/first-person-fairness-in-chatbots.pdf', 'release_date': '2024-10-15'}
| 3
| false
|
OpenAI
|
gpt-4
| null |
mini
| null |
Industry
|
closed
|
North America
|
United States
|
first-party-or-cooperative-evals_50
| true
| 3
| 2,024
|
{'title': 'Evaluating fairness in ChatGPT', 'url': 'https://cdn.openai.com/papers/first-person-fairness-in-chatbots.pdf', 'release_date': '2024-10-15'}
| 3
| false
|
OpenAI
|
gpt-3.5
| null | null |
turbo
|
Industry
|
closed
|
North America
|
United States
|
first-party-or-cooperative-evals_50
| true
| 3
| 2,024
|
{'title': 'Evaluating fairness in ChatGPT', 'url': 'https://cdn.openai.com/papers/first-person-fairness-in-chatbots.pdf', 'release_date': '2024-10-15'}
| 3
| false
|
OpenAI
|
o1
| null |
mini
| null |
Industry
|
closed
|
North America
|
United States
|
first-party-or-cooperative-evals_50
| true
| 3
| 2,024
|
{'title': 'Evaluating fairness in ChatGPT', 'url': 'https://cdn.openai.com/papers/first-person-fairness-in-chatbots.pdf', 'release_date': '2024-10-15'}
| 3
| false
|
OpenAI
|
gpt-4
| null | null |
turbo
|
Industry
|
closed
|
North America
|
United States
|
first-party-or-cooperative-evals_50
| true
| 3
| 2,024
|
{'title': 'Evaluating fairness in ChatGPT', 'url': 'https://cdn.openai.com/papers/first-person-fairness-in-chatbots.pdf', 'release_date': '2024-10-15'}
| 3
| false
|
OpenAI
|
o1
| null |
preview
| null |
Industry
|
closed
|
North America
|
United States
|
first-party-or-cooperative-evals_51
| true
| 2
| 2,024
|
{'title': 'Learning to reason with LLMs', 'url': 'https://openai.com/index/learning-to-reason-with-llms/', 'release_date': '2024-09-12'}
| 1
| false
|
OpenAI
|
gpt-4
| null | null | null |
Industry
|
closed
|
North America
|
United States
|
first-party-or-cooperative-evals_51
| true
| 2
| 2,024
|
{'title': 'Learning to reason with LLMs', 'url': 'https://openai.com/index/learning-to-reason-with-llms/', 'release_date': '2024-09-12'}
| 1
| false
|
OpenAI
|
dall-e
| null | null | null |
Industry
|
closed
|
North America
|
United States
|
first-party-or-cooperative-evals_52
| true
| 3
| 2,024
|
{'title': 'Advancing red teaming with people and AI', 'url': 'https://openai.com/index/advancing-red-teaming-with-people-and-ai/', 'release_date': '2024-11-21'}
| 1
| false
|
OpenAI
|
gpt-4
| null | null |
turbo
|
Industry
|
closed
|
North America
|
United States
|
first-party-or-cooperative-evals_52
| true
| 3
| 2,024
|
{'title': 'Advancing red teaming with people and AI', 'url': 'https://openai.com/index/advancing-red-teaming-with-people-and-ai/', 'release_date': '2024-11-21'}
| 1
| false
|
OpenAI
|
gpt-4
| null | null | null |
Industry
|
closed
|
North America
|
United States
|
first-party-or-cooperative-evals_52
| true
| 3
| 2,024
|
{'title': 'Advancing red teaming with people and AI', 'url': 'https://openai.com/index/advancing-red-teaming-with-people-and-ai/', 'release_date': '2024-11-21'}
| 1
| false
|
OpenAI
|
gpt-4
| null | null |
turbo
|
Industry
|
closed
|
North America
|
United States
|
first-party-or-cooperative-evals_52
| true
| 2
| 2,024
|
{'title': 'Advancing red teaming with people and AI', 'url': 'https://openai.com/index/advancing-red-teaming-with-people-and-ai/', 'release_date': '2024-11-21'}
| 1
| false
|
OpenAI
|
gpt-4
| null | null | null |
Industry
|
closed
|
North America
|
United States
|
first-party-or-cooperative-evals_52
| true
| 2
| 2,024
|
{'title': 'Advancing red teaming with people and AI', 'url': 'https://openai.com/index/advancing-red-teaming-with-people-and-ai/', 'release_date': '2024-11-21'}
| 1
| false
|
OpenAI
|
dall-e
| null | null | null |
Industry
|
closed
|
North America
|
United States
|
first-party-or-cooperative-evals_52
| true
| 2
| 2,024
|
{'title': 'Advancing red teaming with people and AI', 'url': 'https://openai.com/index/advancing-red-teaming-with-people-and-ai/', 'release_date': '2024-11-21'}
| 1
| false
|
OpenAI
|
gpt-5
| null | null | null |
Industry
|
closed
|
North America
|
United States
|
first-party-or-cooperative-evals_53
| true
| 2
| 2,025
|
{'title': 'From hard refusals to safe-completions: toward output-centric safety training', 'url': 'https://cdn.openai.com/pdf/be60c07b-6bc2-4f54-bcee-4141e1d6c69a/gpt-5-safe_completions.pdf', 'release_date': '2025-08-12'}
| 2
| false
|
OpenAI
|
o3
| null | null | null |
Industry
|
closed
|
North America
|
United States
|
first-party-or-cooperative-evals_53
| true
| 2
| 2,025
|
{'title': 'From hard refusals to safe-completions: toward output-centric safety training', 'url': 'https://cdn.openai.com/pdf/be60c07b-6bc2-4f54-bcee-4141e1d6c69a/gpt-5-safe_completions.pdf', 'release_date': '2025-08-12'}
| 2
| false
|
OpenAI
|
o3
| null | null | null |
Industry
|
closed
|
North America
|
United States
|
first-party-or-cooperative-evals_53
| true
| 5
| 2,025
|
{'title': 'From hard refusals to safe-completions: toward output-centric safety training', 'url': 'https://cdn.openai.com/pdf/be60c07b-6bc2-4f54-bcee-4141e1d6c69a/gpt-5-safe_completions.pdf', 'release_date': '2025-08-12'}
| 2
| false
|
OpenAI
|
gpt-5
| null | null | null |
Industry
|
closed
|
North America
|
United States
|
first-party-or-cooperative-evals_53
| true
| 5
| 2,025
|
{'title': 'From hard refusals to safe-completions: toward output-centric safety training', 'url': 'https://cdn.openai.com/pdf/be60c07b-6bc2-4f54-bcee-4141e1d6c69a/gpt-5-safe_completions.pdf', 'release_date': '2025-08-12'}
| 2
| false
|
Lightricks
|
ltx-video
|
13B
| null |
0.9.7-dev
|
Industry
|
open
|
Middle East and North Africa
|
Israel
|
first-party-or-cooperative-evals_54
| false
| 4
| 2,025
|
{'title': 'How Much Power does a SOTA Open Video Model Use?', 'url': 'https://huggingface.co/blog/jdelavande/text-to-video-energy-cost'}
| 2
| false
|
Alibaba
|
wan
|
1.3B
| null |
2.1
|
Industry
|
open
|
East Asia
|
China
|
first-party-or-cooperative-evals_54
| false
| 4
| 2,025
|
{'title': 'How Much Power does a SOTA Open Video Model Use?', 'url': 'https://huggingface.co/blog/jdelavande/text-to-video-energy-cost'}
| 2
| false
|
Alibaba
|
wan
|
14B
| null |
2.1
|
Industry
|
open
|
East Asia
|
China
|
first-party-or-cooperative-evals_54
| false
| 4
| 2,025
|
{'title': 'How Much Power does a SOTA Open Video Model Use?', 'url': 'https://huggingface.co/blog/jdelavande/text-to-video-energy-cost'}
| 2
| false
|
Z.ai
|
cogvideox
|
2B
| null | null |
Industry
|
open
|
East Asia
|
China
|
first-party-or-cooperative-evals_54
| false
| 4
| 2,025
|
{'title': 'How Much Power does a SOTA Open Video Model Use?', 'url': 'https://huggingface.co/blog/jdelavande/text-to-video-energy-cost'}
| 2
| false
|
Genmo
|
mochi
|
10B
|
preview
|
1
|
Industry
|
open
|
North America
|
United States
|
first-party-or-cooperative-evals_54
| false
| 4
| 2,025
|
{'title': 'How Much Power does a SOTA Open Video Model Use?', 'url': 'https://huggingface.co/blog/jdelavande/text-to-video-energy-cost'}
| 2
| false
|
Z.ai
|
cogvideox
|
5B
| null | null |
Industry
|
open
|
East Asia
|
China
|
first-party-or-cooperative-evals_54
| false
| 4
| 2,025
|
{'title': 'How Much Power does a SOTA Open Video Model Use?', 'url': 'https://huggingface.co/blog/jdelavande/text-to-video-energy-cost'}
| 2
| false
|
Alibaba
|
qwen-3
|
30B
| null | null |
Industry
|
open
|
East Asia
|
China
|
first-party-or-cooperative-evals_55
| false
| 4
| 2,025
|
{'title': ""The GPT-OSS models are here… and they're energy-efficient!"", 'url': 'https://huggingface.co/blog/sasha/gpt-oss-energy'}
| 2
| false
|
Microsoft
|
phi-4
|
15B
| null | null |
Industry
|
open
|
North America
|
United States
|
first-party-or-cooperative-evals_55
| false
| 4
| 2,025
|
{'title': ""The GPT-OSS models are here… and they're energy-efficient!"", 'url': 'https://huggingface.co/blog/sasha/gpt-oss-energy'}
| 2
| false
|
Google
|
gemma-3
|
27B
| null | null |
Industry
|
open
|
North America
|
United States
|
first-party-or-cooperative-evals_55
| false
| 4
| 2,025
|
{'title': ""The GPT-OSS models are here… and they're energy-efficient!"", 'url': 'https://huggingface.co/blog/sasha/gpt-oss-energy'}
| 2
| false
|
OpenAI
|
gpt-oss
|
120B
| null | null |
Industry
|
open
|
North America
|
United States
|
first-party-or-cooperative-evals_55
| false
| 4
| 2,025
|
{'title': ""The GPT-OSS models are here… and they're energy-efficient!"", 'url': 'https://huggingface.co/blog/sasha/gpt-oss-energy'}
| 2
| false
|
OpenAI
|
gpt-oss
|
20B
| null | null |
Industry
|
open
|
North America
|
United States
|
first-party-or-cooperative-evals_55
| false
| 4
| 2,025
|
{'title': ""The GPT-OSS models are here… and they're energy-efficient!"", 'url': 'https://huggingface.co/blog/sasha/gpt-oss-energy'}
| 2
| false
|
Cohere
|
command-a
|
110B
| null | null |
Industry
|
open
|
North America
|
Canada
|
first-party-or-cooperative-evals_55
| false
| 4
| 2,025
|
{'title': ""The GPT-OSS models are here… and they're energy-efficient!"", 'url': 'https://huggingface.co/blog/sasha/gpt-oss-energy'}
| 2
| false
|
Meta
|
llama-3.1
|
70B
| null | null |
Industry
|
open
|
North America
|
United States
|
first-party-or-cooperative-evals_55
| false
| 4
| 2,025
|
{'title': ""The GPT-OSS models are here… and they're energy-efficient!"", 'url': 'https://huggingface.co/blog/sasha/gpt-oss-energy'}
| 2
| false
|
OpenAI
|
o3
| null | null | null |
Industry
|
closed
|
North America
|
United States
|
first-party-or-cooperative-evals_56
| true
| 2
| 2,025
|
{'title': 'Safety evaluations hub - Human-sourced jailbreaks', 'url': 'https://openai.com/safety/evaluations-hub/', 'release_date': '2025-08-15'}
| 1
| false
|
OpenAI
|
gpt-4.1
| null |
mini
| null |
Industry
|
closed
|
North America
|
United States
|
first-party-or-cooperative-evals_56
| true
| 2
| 2,025
|
{'title': 'Safety evaluations hub - Human-sourced jailbreaks', 'url': 'https://openai.com/safety/evaluations-hub/', 'release_date': '2025-08-15'}
| 1
| false
|
OpenAI
|
gpt-4.1
| null | null | null |
Industry
|
closed
|
North America
|
United States
|
first-party-or-cooperative-evals_56
| true
| 2
| 2,025
|
{'title': 'Safety evaluations hub - Human-sourced jailbreaks', 'url': 'https://openai.com/safety/evaluations-hub/', 'release_date': '2025-08-15'}
| 1
| false
|
OpenAI
|
gpt-oss
|
20B
| null | null |
Industry
|
open
|
North America
|
United States
|
first-party-or-cooperative-evals_56
| true
| 2
| 2,025
|
{'title': 'Safety evaluations hub - Human-sourced jailbreaks', 'url': 'https://openai.com/safety/evaluations-hub/', 'release_date': '2025-08-15'}
| 1
| false
|
OpenAI
|
gpt-4o
| null | null |
2024-11-20
|
Industry
|
closed
|
North America
|
United States
|
first-party-or-cooperative-evals_56
| true
| 2
| 2,025
|
{'title': 'Safety evaluations hub - Human-sourced jailbreaks', 'url': 'https://openai.com/safety/evaluations-hub/', 'release_date': '2025-08-15'}
| 1
| false
|
OpenAI
|
gpt-4o
| null | null |
2024-05-13
|
Industry
|
closed
|
North America
|
United States
|
first-party-or-cooperative-evals_56
| true
| 2
| 2,025
|
{'title': 'Safety evaluations hub - Human-sourced jailbreaks', 'url': 'https://openai.com/safety/evaluations-hub/', 'release_date': '2025-08-15'}
| 1
| false
|
OpenAI
|
gpt-4
| null |
turbo
| null |
Industry
|
closed
|
North America
|
United States
|
first-party-or-cooperative-evals_56
| true
| 2
| 2,025
|
{'title': 'Safety evaluations hub - Human-sourced jailbreaks', 'url': 'https://openai.com/safety/evaluations-hub/', 'release_date': '2025-08-15'}
| 1
| false
|
OpenAI
|
gpt-oss
|
20B
| null | null |
Industry
|
open
|
North America
|
United States
|
first-party-or-cooperative-evals_56
| true
| 2
| 2,025
|
{'title': 'Safety evaluations hub - Human-sourced jailbreaks', 'url': 'https://openai.com/safety/evaluations-hub/', 'release_date': '2025-08-15'}
| 1
| false
|
OpenAI
|
o3
| null | null | null |
Industry
|
closed
|
North America
|
United States
|
first-party-or-cooperative-evals_56
| true
| 2
| 2,025
|
{'title': 'Safety evaluations hub - Human-sourced jailbreaks', 'url': 'https://openai.com/safety/evaluations-hub/', 'release_date': '2025-08-15'}
| 1
| false
|
OpenAI
|
gpt-4.5
| null | null | null |
Industry
|
closed
|
North America
|
United States
|
first-party-or-cooperative-evals_56
| true
| 2
| 2,025
|
{'title': 'Safety evaluations hub - Human-sourced jailbreaks', 'url': 'https://openai.com/safety/evaluations-hub/', 'release_date': '2025-08-15'}
| 1
| false
|
OpenAI
|
gpt-oss
|
120B
| null | null |
Industry
|
open
|
North America
|
United States
|
first-party-or-cooperative-evals_56
| true
| 2
| 2,025
|
{'title': 'Safety evaluations hub - Human-sourced jailbreaks', 'url': 'https://openai.com/safety/evaluations-hub/', 'release_date': '2025-08-15'}
| 1
| false
|
OpenAI
|
gpt-4.5
| null | null | null |
Industry
|
closed
|
North America
|
United States
|
first-party-or-cooperative-evals_56
| true
| 2
| 2,025
|
{'title': 'Safety evaluations hub - Human-sourced jailbreaks', 'url': 'https://openai.com/safety/evaluations-hub/', 'release_date': '2025-08-15'}
| 1
| false
|
OpenAI
|
gpt-4o
| null | null |
2024-05-13
|
Industry
|
closed
|
North America
|
United States
|
first-party-or-cooperative-evals_56
| true
| 2
| 2,025
|
{'title': 'Safety evaluations hub - Human-sourced jailbreaks', 'url': 'https://openai.com/safety/evaluations-hub/', 'release_date': '2025-08-15'}
| 1
| false
|
OpenAI
|
gpt-4.1
| null | null | null |
Industry
|
closed
|
North America
|
United States
|
first-party-or-cooperative-evals_56
| true
| 2
| 2,025
|
{'title': 'Safety evaluations hub - Human-sourced jailbreaks', 'url': 'https://openai.com/safety/evaluations-hub/', 'release_date': '2025-08-15'}
| 1
| false
|
OpenAI
|
gpt-4
| null |
turbo
| null |
Industry
|
closed
|
North America
|
United States
|
first-party-or-cooperative-evals_56
| true
| 2
| 2,025
|
{'title': 'Safety evaluations hub - Human-sourced jailbreaks', 'url': 'https://openai.com/safety/evaluations-hub/', 'release_date': '2025-08-15'}
| 1
| false
|
OpenAI
|
gpt-4o
| null |
mini
| null |
Industry
|
closed
|
North America
|
United States
|
first-party-or-cooperative-evals_56
| true
| 2
| 2,025
|
{'title': 'Safety evaluations hub - Human-sourced jailbreaks', 'url': 'https://openai.com/safety/evaluations-hub/', 'release_date': '2025-08-15'}
| 1
| false
|
OpenAI
|
gpt-4o
| null | null |
2024-11-20
|
Industry
|
closed
|
North America
|
United States
|
first-party-or-cooperative-evals_56
| true
| 2
| 2,025
|
{'title': 'Safety evaluations hub - Human-sourced jailbreaks', 'url': 'https://openai.com/safety/evaluations-hub/', 'release_date': '2025-08-15'}
| 1
| false
|
OpenAI
|
gpt-4o
| null | null |
2024-05-13
|
Industry
|
closed
|
North America
|
United States
|
first-party-or-cooperative-evals_56
| true
| 5
| 2,025
|
{'title': 'Safety evaluations hub - Human-sourced jailbreaks', 'url': 'https://openai.com/safety/evaluations-hub/', 'release_date': '2025-08-15'}
| 1
| false
|
OpenAI
|
o4
| null |
mini
| null |
Industry
|
closed
|
North America
|
United States
|
first-party-or-cooperative-evals_56
| true
| 5
| 2,025
|
{'title': 'Safety evaluations hub - Human-sourced jailbreaks', 'url': 'https://openai.com/safety/evaluations-hub/', 'release_date': '2025-08-15'}
| 1
| false
|
OpenAI
|
gpt-4.5
| null | null | null |
Industry
|
closed
|
North America
|
United States
|
first-party-or-cooperative-evals_56
| true
| 5
| 2,025
|
{'title': 'Safety evaluations hub - Human-sourced jailbreaks', 'url': 'https://openai.com/safety/evaluations-hub/', 'release_date': '2025-08-15'}
| 1
| false
|
OpenAI
|
o1
| null | null | null |
Industry
|
closed
|
North America
|
United States
|
first-party-or-cooperative-evals_56
| true
| 5
| 2,025
|
{'title': 'Safety evaluations hub - Human-sourced jailbreaks', 'url': 'https://openai.com/safety/evaluations-hub/', 'release_date': '2025-08-15'}
| 1
| false
|
OpenAI
|
gpt-4o
| null | null |
2024-08-16
|
Industry
|
closed
|
North America
|
United States
|
first-party-or-cooperative-evals_56
| true
| 5
| 2,025
|
{'title': 'Safety evaluations hub - Human-sourced jailbreaks', 'url': 'https://openai.com/safety/evaluations-hub/', 'release_date': '2025-08-15'}
| 1
| false
|
OpenAI
|
gpt-4o
| null | null |
2024-11-20
|
Industry
|
closed
|
North America
|
United States
|
first-party-or-cooperative-evals_56
| true
| 5
| 2,025
|
{'title': 'Safety evaluations hub - Human-sourced jailbreaks', 'url': 'https://openai.com/safety/evaluations-hub/', 'release_date': '2025-08-15'}
| 1
| false
|
OpenAI
|
o3
| null | null | null |
Industry
|
closed
|
North America
|
United States
|
first-party-or-cooperative-evals_56
| true
| 5
| 2,025
|
{'title': 'Safety evaluations hub - Human-sourced jailbreaks', 'url': 'https://openai.com/safety/evaluations-hub/', 'release_date': '2025-08-15'}
| 1
| false
|
OpenAI
|
gpt-4.1
| null | null | null |
Industry
|
closed
|
North America
|
United States
|
first-party-or-cooperative-evals_56
| true
| 5
| 2,025
|
{'title': 'Safety evaluations hub - Human-sourced jailbreaks', 'url': 'https://openai.com/safety/evaluations-hub/', 'release_date': '2025-08-15'}
| 1
| false
|
OpenAI
|
o3
| null |
mini
| null |
Industry
|
closed
|
North America
|
United States
|
first-party-or-cooperative-evals_56
| true
| 5
| 2,025
|
{'title': 'Safety evaluations hub - Human-sourced jailbreaks', 'url': 'https://openai.com/safety/evaluations-hub/', 'release_date': '2025-08-15'}
| 1
| false
|
OpenAI
|
gpt-4
| null |
turbo
| null |
Industry
|
closed
|
North America
|
United States
|
first-party-or-cooperative-evals_56
| true
| 5
| 2,025
|
{'title': 'Safety evaluations hub - Human-sourced jailbreaks', 'url': 'https://openai.com/safety/evaluations-hub/', 'release_date': '2025-08-15'}
| 1
| false
|
OpenAI
|
gpt-4.1
| null |
mini
| null |
Industry
|
closed
|
North America
|
United States
|
first-party-or-cooperative-evals_56
| true
| 5
| 2,025
|
{'title': 'Safety evaluations hub - Human-sourced jailbreaks', 'url': 'https://openai.com/safety/evaluations-hub/', 'release_date': '2025-08-15'}
| 1
| false
|
OpenAI
|
gpt-4o
| null |
mini
| null |
Industry
|
closed
|
North America
|
United States
|
first-party-or-cooperative-evals_56
| true
| 5
| 2,025
|
{'title': 'Safety evaluations hub - Human-sourced jailbreaks', 'url': 'https://openai.com/safety/evaluations-hub/', 'release_date': '2025-08-15'}
| 1
| false
|
OpenAI
|
o3
| null | null | null |
Industry
|
closed
|
North America
|
United States
|
first-party-or-cooperative-evals_56
| true
| 2
| 2,025
|
{'title': 'Safety evaluations hub - Human-sourced jailbreaks', 'url': 'https://openai.com/safety/evaluations-hub/', 'release_date': '2025-08-15'}
| 1
| false
|
OpenAI
|
gpt-4o
| null |
mini
| null |
Industry
|
closed
|
North America
|
United States
|
first-party-or-cooperative-evals_56
| true
| 2
| 2,025
|
{'title': 'Safety evaluations hub - Human-sourced jailbreaks', 'url': 'https://openai.com/safety/evaluations-hub/', 'release_date': '2025-08-15'}
| 1
| false
|
OpenAI
|
gpt-4.1
| null | null | null |
Industry
|
closed
|
North America
|
United States
|
first-party-or-cooperative-evals_56
| true
| 2
| 2,025
|
{'title': 'Safety evaluations hub - Human-sourced jailbreaks', 'url': 'https://openai.com/safety/evaluations-hub/', 'release_date': '2025-08-15'}
| 1
| false
|
OpenAI
|
o1
| null | null | null |
Industry
|
closed
|
North America
|
United States
|
first-party-or-cooperative-evals_56
| true
| 2
| 2,025
|
{'title': 'Safety evaluations hub - Human-sourced jailbreaks', 'url': 'https://openai.com/safety/evaluations-hub/', 'release_date': '2025-08-15'}
| 1
| false
|
OpenAI
|
gpt-4.1
| null |
mini
| null |
Industry
|
closed
|
North America
|
United States
|
first-party-or-cooperative-evals_56
| true
| 2
| 2,025
|
{'title': 'Safety evaluations hub - Human-sourced jailbreaks', 'url': 'https://openai.com/safety/evaluations-hub/', 'release_date': '2025-08-15'}
| 1
| false
|
OpenAI
|
gpt-5
| null |
thinking, nano
| null |
Industry
|
closed
|
North America
|
United States
|
first-party-or-cooperative-evals_56
| true
| 2
| 2,025
|
{'title': 'Safety evaluations hub - Human-sourced jailbreaks', 'url': 'https://openai.com/safety/evaluations-hub/', 'release_date': '2025-08-15'}
| 1
| false
|
OpenAI
|
gpt-5
| null |
main, mini
| null |
Industry
|
closed
|
North America
|
United States
|
first-party-or-cooperative-evals_56
| true
| 2
| 2,025
|
{'title': 'Safety evaluations hub - Human-sourced jailbreaks', 'url': 'https://openai.com/safety/evaluations-hub/', 'release_date': '2025-08-15'}
| 1
| false
|
OpenAI
|
gpt-5
| null |
thinking, nano
| null |
Industry
|
closed
|
North America
|
United States
|
first-party-or-cooperative-evals_56
| true
| 2
| 2,025
|
{'title': 'Safety evaluations hub - Human-sourced jailbreaks', 'url': 'https://openai.com/safety/evaluations-hub/', 'release_date': '2025-08-15'}
| 1
| false
|
OpenAI
|
gpt-4o
| null | null |
2024-11-20
|
Industry
|
closed
|
North America
|
United States
|
first-party-or-cooperative-evals_56
| true
| 2
| 2,025
|
{'title': 'Safety evaluations hub - Human-sourced jailbreaks', 'url': 'https://openai.com/safety/evaluations-hub/', 'release_date': '2025-08-15'}
| 1
| false
|
OpenAI
|
gpt-5
| null |
main
| null |
Industry
|
closed
|
North America
|
United States
|
first-party-or-cooperative-evals_56
| true
| 2
| 2,025
|
{'title': 'Safety evaluations hub - Human-sourced jailbreaks', 'url': 'https://openai.com/safety/evaluations-hub/', 'release_date': '2025-08-15'}
| 1
| false
|
OpenAI
|
gpt-4o
| null | null |
2024-08-16
|
Industry
|
closed
|
North America
|
United States
|
first-party-or-cooperative-evals_56
| true
| 2
| 2,025
|
{'title': 'Safety evaluations hub - Human-sourced jailbreaks', 'url': 'https://openai.com/safety/evaluations-hub/', 'release_date': '2025-08-15'}
| 1
| false
|
OpenAI
|
gpt-4o
| null | null |
2024-08-16
|
Industry
|
closed
|
North America
|
United States
|
first-party-or-cooperative-evals_56
| true
| 2
| 2,025
|
{'title': 'Safety evaluations hub - Human-sourced jailbreaks', 'url': 'https://openai.com/safety/evaluations-hub/', 'release_date': '2025-08-15'}
| 1
| false
|
OpenAI
|
gpt-5
| null |
thinking, mini
| null |
Industry
|
closed
|
North America
|
United States
|
first-party-or-cooperative-evals_56
| true
| 2
| 2,025
|
{'title': 'Safety evaluations hub - Human-sourced jailbreaks', 'url': 'https://openai.com/safety/evaluations-hub/', 'release_date': '2025-08-15'}
| 1
| false
|
OpenAI
|
gpt-4.1
| null |
mini
| null |
Industry
|
closed
|
North America
|
United States
|
first-party-or-cooperative-evals_56
| true
| 2
| 2,025
|
{'title': 'Safety evaluations hub - Human-sourced jailbreaks', 'url': 'https://openai.com/safety/evaluations-hub/', 'release_date': '2025-08-15'}
| 1
| false
|
OpenAI
|
gpt-5
| null |
thinking, mini
| null |
Industry
|
closed
|
North America
|
United States
|
first-party-or-cooperative-evals_56
| true
| 2
| 2,025
|
{'title': 'Safety evaluations hub - Human-sourced jailbreaks', 'url': 'https://openai.com/safety/evaluations-hub/', 'release_date': '2025-08-15'}
| 1
| false
|
OpenAI
|
gpt-4o
| null | null |
2024-05-13
|
Industry
|
closed
|
North America
|
United States
|
first-party-or-cooperative-evals_56
| true
| 2
| 2,025
|
{'title': 'Safety evaluations hub - Human-sourced jailbreaks', 'url': 'https://openai.com/safety/evaluations-hub/', 'release_date': '2025-08-15'}
| 1
| false
|
OpenAI
|
gpt-5
| null |
main
| null |
Industry
|
closed
|
North America
|
United States
|
first-party-or-cooperative-evals_56
| true
| 2
| 2,025
|
{'title': 'Safety evaluations hub - Human-sourced jailbreaks', 'url': 'https://openai.com/safety/evaluations-hub/', 'release_date': '2025-08-15'}
| 1
| false
|
OpenAI
|
o4
| null |
mini
| null |
Industry
|
closed
|
North America
|
United States
|
first-party-or-cooperative-evals_56
| true
| 2
| 2,025
|
{'title': 'Safety evaluations hub - Human-sourced jailbreaks', 'url': 'https://openai.com/safety/evaluations-hub/', 'release_date': '2025-08-15'}
| 1
| false
|
OpenAI
|
o3
| null | null | null |
Industry
|
closed
|
North America
|
United States
|
first-party-or-cooperative-evals_56
| true
| 2
| 2,025
|
{'title': 'Safety evaluations hub - Human-sourced jailbreaks', 'url': 'https://openai.com/safety/evaluations-hub/', 'release_date': '2025-08-15'}
| 1
| false
|
OpenAI
|
gpt-5
| null |
thinking
| null |
Industry
|
closed
|
North America
|
United States
|
first-party-or-cooperative-evals_56
| true
| 2
| 2,025
|
{'title': 'Safety evaluations hub - Human-sourced jailbreaks', 'url': 'https://openai.com/safety/evaluations-hub/', 'release_date': '2025-08-15'}
| 1
| false
|
OpenAI
|
gpt-5
| null |
thinking
| null |
Industry
|
closed
|
North America
|
United States
|
first-party-or-cooperative-evals_56
| true
| 2
| 2,025
|
{'title': 'Safety evaluations hub - Human-sourced jailbreaks', 'url': 'https://openai.com/safety/evaluations-hub/', 'release_date': '2025-08-15'}
| 1
| false
|
OpenAI
|
gpt-4
| null |
turbo
| null |
Industry
|
closed
|
North America
|
United States
|
first-party-or-cooperative-evals_56
| true
| 2
| 2,025
|
{'title': 'Safety evaluations hub - Human-sourced jailbreaks', 'url': 'https://openai.com/safety/evaluations-hub/', 'release_date': '2025-08-15'}
| 1
| false
|
OpenAI
|
gpt-4.5
| null | null | null |
Industry
|
closed
|
North America
|
United States
|
first-party-or-cooperative-evals_56
| true
| 2
| 2,025
|
{'title': 'Safety evaluations hub - Human-sourced jailbreaks', 'url': 'https://openai.com/safety/evaluations-hub/', 'release_date': '2025-08-15'}
| 1
| false
|
OpenAI
|
o4
| null |
mini
| null |
Industry
|
closed
|
North America
|
United States
|
first-party-or-cooperative-evals_56
| true
| 2
| 2,025
|
{'title': 'Safety evaluations hub - Human-sourced jailbreaks', 'url': 'https://openai.com/safety/evaluations-hub/', 'release_date': '2025-08-15'}
| 1
| false
|
OpenAI
|
o3
| null | null | null |
Industry
|
closed
|
North America
|
United States
|
first-party-or-cooperative-evals_56
| true
| 2
| 2,025
|
{'title': 'Safety evaluations hub - Human-sourced jailbreaks', 'url': 'https://openai.com/safety/evaluations-hub/', 'release_date': '2025-08-15'}
| 1
| false
|
OpenAI
|
o4
| null |
mini
| null |
Industry
|
closed
|
North America
|
United States
|
first-party-or-cooperative-evals_56
| true
| 2
| 2,025
|
{'title': 'Safety evaluations hub - Human-sourced jailbreaks', 'url': 'https://openai.com/safety/evaluations-hub/', 'release_date': '2025-08-15'}
| 1
| false
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.