Merge pull request #158 from stanford-crfm/jonathan/0216-weekly-assets

update changes

assets/01ai.yaml

@@ -19,7 +19,8 @@
     diversity of responses.
   access: open
   license:
-    explanation: Model license can be found at https://huggingface.co/01-ai/Yi-34B/blob/main/LICENSE. Code license is under Apache 2.0
+    explanation: Model license can be found at https://huggingface.co/01-ai/Yi-34B/blob/main/LICENSE.
+      Code license is under Apache 2.0
     value: custom
   intended_uses: ''
   prohibited_uses: none
@@ -46,7 +47,8 @@
   quality_control: unknown
   access: open
   license:
-    explanation: Model license can be found at https://huggingface.co/01-ai/Yi-VL-34B/blob/main/LICENSE. Code license is under Apache 2.0
+    explanation: Model license can be found at https://huggingface.co/01-ai/Yi-VL-34B/blob/main/LICENSE.
+      Code license is under Apache 2.0
     value: custom
   intended_uses: ''
   prohibited_uses: ''

assets/ai21.yaml

@@ -298,12 +298,14 @@
 - type: model
   name: Jamba
   organization: AI21 Labs
-  description: Jamba is a state-of-the-art, hybrid SSM-Transformer LLM. Jamba is the world’s first production-grade Mamba based model.
+  description: Jamba is a state-of-the-art, hybrid SSM-Transformer LLM. Jamba is
+    the world’s first production-grade Mamba based model.
   created_date: 2024-03-28
   url: https://www.ai21.com/blog/announcing-jamba
   model_card: https://huggingface.co/ai21labs/Jamba-v0.1
   modality: text; text
-  analysis: Jamba outperforms or matches other state-of-the-art models in its size class on a wide range of benchmarks.
+  analysis: Jamba outperforms or matches other state-of-the-art models in its size
+    class on a wide range of benchmarks.
   size: 52B parameters (sparse)
   dependencies: []
   training_emissions: unknown

assets/alibaba.yaml

@@ -74,7 +74,8 @@
       \ repository]](https://huggingface.co/Qwen)\n"
     value: open
   license:
-    explanation: Model license can be found at https://github.com/QwenLM/Qwen/blob/main/Tongyi%20Qianwen%20LICENSE%20AGREEMENT. Code license is under Apache 2.0
+    explanation: Model license can be found at https://github.com/QwenLM/Qwen/blob/main/Tongyi%20Qianwen%20LICENSE%20AGREEMENT.
+      Code license is under Apache 2.0
     value: custom
   intended_uses: ''
   prohibited_uses: ''
@@ -87,12 +88,15 @@
 - type: model
   name: Qwen 1.5
   organization: Qwen AI
-  description: Qwen 1.5 is the next iteration in their Qwen series, consisting of Transformer-based large language models pretrained on a large volume of data, including web texts, books, codes, etc.
+  description: Qwen 1.5 is the next iteration in their Qwen series, consisting of
+    Transformer-based large language models pretrained on a large volume of data,
+    including web texts, books, codes, etc.
   created_date: 2024-02-04
   url: https://qwenlm.github.io/blog/qwen1.5/
   model_card: https://huggingface.co/Qwen/Qwen1.5-72B
   modality: text; text
-  analysis: Evaluated on MMLU, C-Eval, GSM8K, MATH, HumanEval, MBPP, BBH, CMMLU, all standard English and Chinese benchmarks.
+  analysis: Evaluated on MMLU, C-Eval, GSM8K, MATH, HumanEval, MBPP, BBH, CMMLU,
+    all standard English and Chinese benchmarks.
   size: 72B parameters (dense)
   dependencies: []
   training_emissions: unknown

assets/anthropic.yaml

@@ -573,22 +573,28 @@
 - type: model
   name: Claude 3
   organization: Anthropic
-  description: The Claude 3 model family is a collection of models which sets new industry benchmarks across a wide range of cognitive tasks.
+  description: The Claude 3 model family is a collection of models which sets new
+    industry benchmarks across a wide range of cognitive tasks.
   created_date: 2024-03-04
   url: https://www.anthropic.com/news/claude-3-family
   model_card: https://www-cdn.anthropic.com/de8ba9b01c9ab7cbabf5c33b80b7bbc618857627/Model_Card_Claude_3.pdf
   modality: image, text; text
-  analysis: Evaluated on reasoning, math, coding, reading comprehension, and question answering, outperforming GPT-4 on standard benchmarks.
+  analysis: Evaluated on reasoning, math, coding, reading comprehension, and question
+    answering, outperforming GPT-4 on standard benchmarks.
   size: unknown
   dependencies: []
   training_emissions: unknown
   training_time: unknown
   training_hardware: unknown
-  quality_control: Pre-trained on diverse dataset and aligned with Constitutional AI technique.
+  quality_control: Pre-trained on diverse dataset and aligned with Constitutional
+    AI technique.
   access: limited
   license: unknown
-  intended_uses: Claude models excel at open-ended conversation and collaboration on ideas, and also perform exceptionally well in coding tasks and when working with text - whether searching, writing, editing, outlining, or summarizing.
-  prohibited_uses: Prohibited uses include, but are not limited to, political campaigning or lobbying, surveillance, social scoring, criminal justice decisions, law enforcement, and decisions related to financing, employment, and housing.
+  intended_uses: Claude models excel at open-ended conversation and collaboration
+    on ideas, and also perform exceptionally well in coding tasks and when working
+    with text - whether searching, writing, editing, outlining, or summarizing.
+  prohibited_uses: Prohibited uses include, but are not limited to, political campaigning
+    or lobbying, surveillance, social scoring, criminal justice decisions, law enforcement,
+    and decisions related to financing, employment, and housing.
   monitoring: ''
   feedback: none
-

assets/apple.yaml

@@ -2,12 +2,14 @@
 - type: model
   name: MM1
   organization: Apple
-  description: MM1 is a family of multimodal models, including both dense variants up to 30B and mixture-of-experts (MoE) variants up to 64B.
+  description: MM1 is a family of multimodal models, including both dense variants
+    up to 30B and mixture-of-experts (MoE) variants up to 64B.
   created_date: 2024-03-16
   url: https://arxiv.org/pdf/2403.09611.pdf
   model_card: none
   modality: image, text; text
-  analysis: Evaluated on image captioning and visual question answering across many benchmarks.
+  analysis: Evaluated on image captioning and visual question answering across many
+    benchmarks.
   size: 30B parameters (dense)
   dependencies: []
   training_emissions: unknown

assets/avignon.yaml

@@ -2,12 +2,15 @@
 - type: model
   name: BioMistral
   organization: Avignon University, Nantes University
-  description: BioMistral is an open-source Large Language Model tailored for the biomedical domain, utilizing Mistral as its foundation model and further pre-trained on PubMed Central.
+  description: BioMistral is an open-source Large Language Model tailored for the
+    biomedical domain, utilizing Mistral as its foundation model and further pre-trained
+    on PubMed Central.
   created_date: 2024-02-15
   url: https://arxiv.org/pdf/2402.10373.pdf
   model_card: https://huggingface.co/BioMistral/BioMistral-7B
   modality: text; text
-  analysis: BioMistral was evaluated on a benchmark comprising 10 established medical question-answering (QA) tasks in English and seven other languages.
+  analysis: BioMistral was evaluated on a benchmark comprising 10 established medical
+    question-answering (QA) tasks in English and seven other languages.
   size: 7B parameters (dense)
   dependencies: [Mistral, PubMed Central]
   training_emissions: unknown
@@ -16,7 +19,9 @@
   quality_control: ''
   access: open
   license: Apache 2.0
-  intended_uses: Research in the biomedical domain, especially for medical question-answering tasks.
-  prohibited_uses: Prohibited from deploying in production environments for natural language generation or any professional health and medical purposes.
+  intended_uses: Research in the biomedical domain, especially for medical question-answering
+    tasks.
+  prohibited_uses: Prohibited from deploying in production environments for natural
+    language generation or any professional health and medical purposes.
   monitoring: ''
-  feedback: https://huggingface.co/BioMistral/BioMistral-7B/discussions
+  feedback: https://huggingface.co/BioMistral/BioMistral-7B/discussions

assets/baai.yaml

@@ -147,12 +147,14 @@
 - type: model
   name: EVA-CLIP
   organization: Beijing Academy of Artificial Intelligence, Tsinghua University
-  description: As of release, EVA-CLIP is the largest and most powerful open-source CLIP model to date, with 18 billion parameters.
+  description: As of release, EVA-CLIP is the largest and most powerful open-source
+    CLIP model to date, with 18 billion parameters.
   created_date: 2024-02-06
   url: https://arxiv.org/pdf/2402.04252.pdf
   model_card: https://huggingface.co/BAAI/EVA-CLIP-8B-448
   modality: image, text; text
-  analysis: Evaluated on zero-shot classification performance across multiple image classification benchmarks.
+  analysis: Evaluated on zero-shot classification performance across multiple image
+    classification benchmarks.
   size: 18B parameters (dense)
   dependencies: [CLIP]
   training_emissions: unknown

assets/beitech.yaml

@@ -43,4 +43,3 @@
   prohibited_uses: ''
   monitoring: unknokwn
   feedback: https://huggingface.co/GeneZC/MiniMA-3B/discussions
-

assets/bigcode.yaml

@@ -74,7 +74,9 @@
 - type: model
   name: StarCoder2
   organization: BigCode
-  description: A 15 billion parameter language model trained on 600+ programming languages from The Stack v2. The training was carried out using the Fill-in-the-Middle objective on 4+ trillion tokens.
+  description: A 15 billion parameter language model trained on 600+ programming
+    languages from The Stack v2. The training was carried out using the Fill-in-the-Middle
+    objective on 4+ trillion tokens.
   created_date: 2024-02-28
   url: https://www.servicenow.com/company/media/press-room/huggingface-nvidia-launch-starcoder2.html
   model_card: https://huggingface.co/bigcode/starcoder2-15b
@@ -85,10 +87,14 @@
   training_emissions: unknown
   training_time: unknown
   training_hardware: 1024 x H100 GPUs
-  quality_control: The model was filtered for permissive licenses and code with no license only. A search index is provided to identify where generated code came from to apply the proper attribution.
+  quality_control: The model was filtered for permissive licenses and code with
+    no license only. A search index is provided to identify where generated code
+    came from to apply the proper attribution.
   access: open
   license: BigCode OpenRail-M
-  intended_uses: Intended to generate code snippets from given context, but not for writing actual functional code directly.
-  prohibited_uses: Should not be used as a way to write fully functioning code without modification or verification.
+  intended_uses: Intended to generate code snippets from given context, but not
+    for writing actual functional code directly.
+  prohibited_uses: Should not be used as a way to write fully functioning code without
+    modification or verification.
   monitoring: unknown
   feedback: https://huggingface.co/bigcode/starcoder2-15b/discussions

assets/bytedance.yaml

@@ -25,12 +25,16 @@
 - type: model
   name: SDXL-Lightning
   organization: ByteDance
-  description: SDXL-Lightning is a lightning-fast text-to-image generation model. It can generate high-quality 1024px images in a few steps. The models are distilled from stabilityai/stable-diffusion-xl-base-1.0. This repository contains checkpoints for 1-step, 2-step, 4-step, and 8-step distilled models.
+  description: SDXL-Lightning is a lightning-fast text-to-image generation model.
+    It can generate high-quality 1024px images in a few steps. The models are distilled
+    from stabilityai/stable-diffusion-xl-base-1.0. This repository contains checkpoints
+    for 1-step, 2-step, 4-step, and 8-step distilled models.
   created_date: 2024-02-21
   url: https://arxiv.org/pdf/2402.13929.pdf
   model_card: https://huggingface.co/ByteDance/SDXL-Lightning
   modality: text; image
-  analysis: Evaluated via qualitative comparison relative to other SoTA image generation models.
+  analysis: Evaluated via qualitative comparison relative to other SoTA image generation
+    models.
   size: unknown
   dependencies: [Stable Diffusion XL]
   training_emissions: unknown
@@ -39,7 +43,9 @@
   quality_control: unknown
   access: open
   license: OpenRail++
-  intended_uses: The model can be used for fast, high-quality text-to-image generation. It supports 1-step, 2-step, 4-step, and 8-step distilled models which provide varying generation quality.
+  intended_uses: The model can be used for fast, high-quality text-to-image generation.
+    It supports 1-step, 2-step, 4-step, and 8-step distilled models which provide
+    varying generation quality.
   prohibited_uses: unknown
   monitoring: unknown
   feedback: https://huggingface.co/ByteDance/SDXL-Lightning/discussions

assets/cagliostro.yaml

@@ -2,7 +2,10 @@
 - type: model
   name: Animagine XL 3.1
   organization: Cagliostro Research Lab
-  description: An open-source, anime-themed text-to-image model enhanced to generate higher quality anime-style images with a broader range of characters from well-known anime series, an optimized dataset, and new aesthetic tags for better image creation.
+  description: An open-source, anime-themed text-to-image model enhanced to generate
+    higher quality anime-style images with a broader range of characters from well-known
+    anime series, an optimized dataset, and new aesthetic tags for better image
+    creation.
   created_date: 2024-03-18
   url: https://cagliostrolab.net/posts/animagine-xl-v31-release
   model_card: https://huggingface.co/cagliostrolab/animagine-xl-3.1
@@ -13,10 +16,14 @@
   training_emissions: unknown
   training_time: Approximately 15 days, totaling over 350 GPU hours.
   training_hardware: 2x A100 80GB GPUs
-  quality_control: The model undergoes pretraining, first stage finetuning, and second stage finetuning for refining and improving aspects such as hand and anatomy rendering.
+  quality_control: The model undergoes pretraining, first stage finetuning, and
+    second stage finetuning for refining and improving aspects such as hand and
+    anatomy rendering.
   access: open
   license: Fair AI Public License 1.0-SD
-  intended_uses: Generating high-quality anime images from textual prompts. Useful for anime fans, artists, and content creators.
-  prohibited_uses: Not suitable for creating realistic photos or for users who expect high-quality results from short or simple prompts.
+  intended_uses: Generating high-quality anime images from textual prompts. Useful
+    for anime fans, artists, and content creators.
+  prohibited_uses: Not suitable for creating realistic photos or for users who expect
+    high-quality results from short or simple prompts.
   monitoring: unknown
   feedback: https://huggingface.co/cagliostrolab/animagine-xl-3.1/discussions

assets/causallm.yaml

@@ -2,7 +2,8 @@
 - type: model
   name: CausalLM
   organization: CausalLM
-  description: CausalLM is an LLM based on the model weights of Qwen and trained on a model architecture identical to LLaMA 2.
+  description: CausalLM is an LLM based on the model weights of Qwen and trained
+    on a model architecture identical to LLaMA 2.
   created_date: 2023-10-21
   url: https://huggingface.co/CausalLM/14B
   model_card: https://huggingface.co/CausalLM/14B
@@ -15,8 +16,9 @@
   training_hardware: unknown
   quality_control: ''
   access: open
-  license: 
-    explanation: can be found at https://github.com/rpherrera/WTFPL (HuggingFace lists this to be the license)
+  license:
+    explanation: can be found at https://github.com/rpherrera/WTFPL (HuggingFace
+      lists this to be the license)
     value: WTFPL
   intended_uses: ''
   prohibited_uses: ''

assets/cerebras.yaml

@@ -117,12 +117,14 @@
 - type: model
   name: Bittensor Language Model
   organization: Cerebras
-  description: Bittensor Language Model is a 3 billion parameter language model with an 8k context length trained on 627B tokens of SlimPajama.
+  description: Bittensor Language Model is a 3 billion parameter language model
+    with an 8k context length trained on 627B tokens of SlimPajama.
   created_date: 2023-07-24
   url: https://www.cerebras.net/blog/btlm-3b-8k-7b-performance-in-a-3-billion-parameter-model/
   model_card: https://huggingface.co/cerebras/btlm-3b-8k-base
   modality: text; text
-  analysis: Evaluated on standard LLM benchmarks in comparison to similar-sized models.
+  analysis: Evaluated on standard LLM benchmarks in comparison to similar-sized
+    models.
   size: 3B parameters (dense)
   dependencies: [SlimPajama]
   training_emissions: unknown
@@ -138,7 +140,8 @@
 - type: dataset
   name: SlimPajama
   organization: Cerebras
-  description: As of release, SlimPajama is the largest extensively deduplicated, multi-corpora, open-source dataset for training large language models.
+  description: As of release, SlimPajama is the largest extensively deduplicated,
+    multi-corpora, open-source dataset for training large language models.
   created_date: 2023-06-09
   url: https://huggingface.co/datasets/cerebras/SlimPajama-627B
   datasheet: https://huggingface.co/datasets/cerebras/SlimPajama-627B

assets/cmu.yaml

@@ -34,7 +34,8 @@
 - type: model
   name: Moment
   organization: Carnegie Mellon University, University of Pennsylvania
-  description: Moment is a family of open-source foundation models for general-purpose time-series analysis.
+  description: Moment is a family of open-source foundation models for general-purpose
+    time-series analysis.
   created_date: 2024-02-06
   url: https://arxiv.org/pdf/2402.03885.pdf
   model_card: none

assets/cognition.yaml

@@ -7,7 +7,8 @@
   url: https://www.cognition-labs.com/introducing-devin
   model_card: none
   modality: text; code
-  analysis: Evaluated on SWE-Bench, a challenging software engineering benchmark, where Devin outperforms major state of the art models unassisted.
+  analysis: Evaluated on SWE-Bench, a challenging software engineering benchmark,
+    where Devin outperforms major state of the art models unassisted.
   size: unknown
   dependencies: []
   training_emissions: unknown

assets/cognitive.yaml

@@ -25,9 +25,11 @@
 - type: model
   name: WizardLM Uncensored
   organization: Cognitive Computations
-  description: WizardLM Uncensored is WizardLM trained with a subset of the dataset - responses that contained alignment / moralizing were removed.
+  description: WizardLM Uncensored is WizardLM trained with a subset of the dataset
+    - responses that contained alignment / moralizing were removed.
   created_date:
-    explanation: release date is not published; estimated to be sometime in either May or June 2023.
+    explanation: release date is not published; estimated to be sometime in either
+      May or June 2023.
     value: 2023-06-01
   url: https://huggingface.co/cognitivecomputations/WizardLM-30B-Uncensored
   model_card: https://huggingface.co/cognitivecomputations/WizardLM-30B-Uncensored
@@ -45,4 +47,3 @@
   prohibited_uses: ''
   monitoring: unknown
   feedback: https://huggingface.co/cognitivecomputations/WizardLM-30B-Uncensored/discussions
-