Type | Name | Organization | Created date | Size | Access | License | Dependencies |
---|---|---|---|---|---|---|---|
model | MathCoder | Shanghai AI Laboratory | Oct 5, 2023 | 70B parameters (dense) | open | unknown | GPT-4 | LLaMA 2 |
model | RT-2-X | Open X-Embodiment, Google Deepmind | Oct 3, 2023 | 55B parameters (dense) | closed | unknown | Open X-Embodiment dataset | ViT (unknown size) | UL2 |
model | RT-1-X | Open X-Embodiment, Google Deepmind | Oct 3, 2023 | 35M parameters (dense) | open | unknown | Open X-Embodiment dataset | ImageNet EfficientNet | USE |
dataset | Open X-Embodiment dataset | Open X-Embodiment | Oct 3, 2023 | 160K tasks | open | unknown | |
model | GAIA-1 | Wayve | Sep 29, 2023 | 9B parameters (dense) | closed | unknown | |
model | Emu | Meta | Sep 27, 2023 | 1.5B parameters (dense) | open | unknown | CLIP | T5 |
model | MAmmoTH | Ohio State University | Sep 11, 2023 | 34B parameters (dense) | open | Apache 2.0 | MathInstruct | LLaMA | Code LLaMA |
application | Watsonx.ai | IBM | Sep 7, 2023 | n/a | limited | ||
model | Persimmon | Adept | Sep 7, 2023 | 8B parameters (dense) | open | Apache 2.0 | |
model | Falcon-180B | UAE Technology Innovation Institute | Sep 6, 2023 | 180B parameters (dense) | open | unknown | RefinedWeb |
application | ChatGPT Enterprise | OpenAI | Aug 28, 2023 | n/a | limited | custom | GPT-4 |
model | WizardCoder | Microsoft | Aug 26, 2023 | 34B parameters (dense) | open | Apache 2.0 | Evol-Instruct | Alpaca dataset | StarCoder |
model | Llama-2-7B-32K-Instruct | Together | Aug 18, 2023 | 7B parameters (dense) | open | Apache 2.0 | BookSum dataset | MQA dataset | Together API | LLaMA 2 |
dataset | Dolma | AI2 | Aug 18, 2023 | 3T tokens | open | custom | |
model | Platypus | Boston University | Aug 14, 2023 | 13B parameters (dense) | open | CC by-NC-SA 4.0 | LLaMA 2 | Platypus curated dataset |
model | Prithvi | IBM | Aug 3, 2023 | 100M parameters (dense) | open | Apache 2.0 | NASA HLS data |
model | MusicGen | Meta | Aug 2, 2023 | 3.3B parameters (dense) | open | MIT | Meta Music Initative Sound Collection | Shutterstock music collection | Pond5 music collection |
model | AudioGen | Meta | Aug 2, 2023 | 1.5B parameters (dense) | open | MIT | AudioSet | BBC sound effects | AudioCaps | Clotho v2 | VGG-Sound | FSD50K | Free To Use Sounds | Sonniss Game Effects | WeSoundEffects | Paramount Motion – Odeon Cinematic Sound Effects |
dataset | LP-MusicCaps | South Korea Graduate School of Culture Technology | Jul 31, 2023 | 2.2M captions paired with 0.5M audio clips | open | CC BY 4.0 | MusicCaps | Million Song Dataset | Magnatagtune |
model | RT-2 | DeepMind | Jul 28, 2023 | 55B parameters (dense) | open | unknown | PaLI-X | PaLM-E | RT-2 action tokens |
application | Stable Diffusion XL | Stability AI | Jul 26, 2023 | n/a | limited | MIT | |
model | Med-PaLM Multimodal | Jul 26, 2023 | 562B parameters (dense) | closed | unknown | PaLM-E | MultiMedBench | |
model | LLaMA 2 | Meta | Jul 18, 2023 | 70B parameters (dense) | open | custom | |
model | Claude 2 | Anthropic | Jul 11, 2023 | open | Claude human feedback data | Unknown licensed third party datasets | ||
model | Inflection-1 | Inflection AI | Jun 22, 2023 | unknown | limited | unknown | |
model | h2oGPT | H2O AI | Jun 16, 2023 | 20B parameters (dense) | open | Apache 2.0 | GPT-NeoX | H2O AI OpenAssistant | h2oGPT Repositories |
model | Voicebox | Meta | Jun 16, 2023 | 330M parameters (dense) | closed | ||
model | Falcon-40B | UAE Technology Innovation Institute | Jun 14, 2023 | 40B parameters (dense) | open | Apache 2.0 | RefinedWeb |
model | CORGI | Stanford | Jun 12, 2023 | 124M parameters (dense) | open | MIT | GPT-2 | BABEL | text-davinci-003 |
dataset | Multimodal C4 | AI2 | Jun 9, 2023 | 43B English tokens with 101.2M documents and 571M images | open | MIT | C4 |
model | Google Joint SLM | Jun 8, 2023 | unknown | open | CTC blank-filtering | Speech2Text adapter | ||
dataset | RefinedWeb | UAE Technology Innovation Institute | Jun 1, 2023 | 600B tokens | open | custom | |
model | Pythia | Eleuther AI | May 31, 2023 | 12B parameters (dense) | open | Apache 2.0 | The Pile |
application | Transformify Automate | Transformify | May 30, 2023 | n/a | open | GPT-4 | |
model | Lego-MT | Shanghai AI Laboratory | May 29, 2023 | 1.2B parameters (dense) | open | OPUS | |
model | BigTrans | Institute of Automation Chinese Academy of Sciences | May 29, 2023 | 13B parameters (dense) | open | Apache 2.0 | LLaMA | CLUE | BigTrans parallel dataset |
model | BiomedGPT | Lehigh University | May 26, 2023 | 472M parameters (dense) | open | Apache 2.0 | GPT-style autoregressive decoder | BiomedGPT biomedical datasets |
dataset | SODA | AI2 | May 24, 2023 | 1.5M dialogues | open | CC BY 4.0 | |
model | Gorilla | Berkeley | May 24, 2023 | 7B parameters (dense) | open | Apache 2.0 | LLaMA | Gorilla document retriever |
model | COSMO | AI2 | May 24, 2023 | 11B parameters (dense) | open | SODA | ProsocialDialog | T5 | |
model | Guanaco | University of Washington | May 23, 2023 | 33B parameters (dense) | open | MIT | QLoRA | OASST1 |
model | GOAT | National University of Singapore | May 23, 2023 | 7B parameters (dense) | open | Apache 2.0 | LLaMA | GOAT dataset |
model | PaLM 2 | May 10, 2023 | unknown | open | PaLM 2 dataset | ||
model | StarCoder | BigCode | May 9, 2023 | 15.5B parameters (dense) | open | Apache 2.0 | The Stack |
application | Portkey | Portkey | May 6, 2023 | n/a | open | ||
model | Otter | Nanyang Technological University | May 5, 2023 | 1.3B parameters (dense) | open | MIT | MIMIC-IT | OpenFlamingo |
model | MPT | Mosaic | May 5, 2023 | 7B parameters (dense) | open | Apache 2.0 | RedPajama-Data | C4 | The Stack | Multimodal C4 |
model | OpenLLaMA | Berkeley | May 3, 2023 | 17B parameters (dense) | open | Apache 2.0 | RedPajama |
application | Pi | Inflection AI | May 2, 2023 | n/a | limited | unknown | Inflection-1 |
application | Nextdoor Assistant | Nextdoor | May 2, 2023 | n/a | open | unknown | ChatGPT |
model | DeepFloyd IF | Stability AI | Apr 28, 2023 | 4.3B parameters (dense) | open | custom | LAION-5B |
application | ARES | Faraday Lab | Apr 26, 2023 | n/a | open | unknown | Stable Diffusion |
model | WizardLM | Microsoft | Apr 24, 2023 | 7B parameters (dense) | open | Apache 2.0 | LLaMA | Evol-Instruct | Alpaca dataset |
model | StableLM | Stability AI | Apr 20, 2023 | 7B parameters (dense) | open | Apache 2.0 | StableLM-Alpha dataset | Alpaca dataset | gpt4all dataset | ShareGPT52K dataset | Dolly dataset | HH dataset |
model | Bark | Suno | Apr 20, 2023 | open | MIT | AudioLM | |
application | Auto-GPT | Auto-GPT | Apr 16, 2023 | n/a | open | MIT | GPT-4 API |
application | Bedrock | Amazon | Apr 13, 2023 | n/a | limited | unknown | Jurassic-2 | Claude | Stable Diffusion | Amazon Titan | Claude 2 | Cohere Command |
model | SAM | Meta | Apr 5, 2023 | unknown | open | Apache 2.0 | SA-1B |
dataset | SA-1B | Meta | Apr 5, 2023 | 11M images, 1.1B mask annotations | open | SA-1B Dataset Research License | |
model | Koala | Berkeley | Apr 3, 2023 | 13B parameters (dense) | open | Apache 2.0 | LLaMA | web-scraped dialogue data |
model | Camel | Writer | Apr 1, 2023 | 5B parameters (dense) | open | Apache 2.0 | Palmyra | Camel dataset |
model | Vicuna | LMSYS | Mar 30, 2023 | 13B parameters (dense) | open | Apache 2.0 | LLaMA | ShareGPT conversations data |
dataset | FinPile | Bloomberg | Mar 30, 2023 | 363B tokens | closed | unknown | |
model | BloombergGPT | Bloomberg | Mar 30, 2023 | 50B parameters (dense) | closed | unknown | FinPile | The Pile | C4 | Wikipedia |
model | OpenFlamingo | LAION | Mar 28, 2023 | 9B parameters (dense) | open | MIT | LLaMA | CLIP |
application | Microsoft Security Copilot | Microsoft | Mar 28, 2023 | n/a | limited | custom | GPT-4 | Microsoft security-specific model |
model | Cerebras-GPT | Cerebras | Mar 28, 2023 | 13B parameters (dense) | open | Apache 2.0 | The Pile |
model | Dolly | Databricks | Mar 24, 2023 | 6B parameters (dense) | open | Apache 2.0 | GPT-J | Alpaca dataset |
application | Cformers | Nolano | Mar 19, 2023 | n/a | limited | MIT | |
application | Microsoft Business Chat | Microsoft | Mar 16, 2023 | n/a | limited | custom | Microsoft 365 Copilot |
application | Microsoft 365 Copilot | Microsoft | Mar 16, 2023 | n/a | limited | custom | GPT-4 API |
dataset | Conformer-1 dataset | AssemblyAI | Mar 15, 2023 | 650K hours audio (60TB) | closed | unknown | |
application | Conformer-1 API | AssemblyAI | Mar 15, 2023 | n/a | open | custom | Conformer-1 |
model | Conformer-1 | AssemblyAI | Mar 15, 2023 | 300M parameters (dense) | limited | unknown | Conformer-1 dataset |
application | Virtual Volunteer | Be My Eyes | Mar 14, 2023 | n/a | limited | unknown | GPT-4 API |
application | PaLM API | Mar 14, 2023 | n/a | limited | unknown | PaLM | |
application | Khanmigo | Khan Academy | Mar 14, 2023 | n/a | limited | unknown | GPT-4 API |
application | GPT-4 API | OpenAI | Mar 14, 2023 | n/a | limited | custom | GPT-4 |
model | GPT-4 | OpenAI | Mar 14, 2023 | unknown | limited | unknown | |
application | Duolingo Role Play | Duolingo | Mar 14, 2023 | n/a | limited | custom | GPT-4 API |
application | Duolingo Max | Duolingo | Mar 14, 2023 | n/a | limited | custom | Duolingo Role Play | Duolingo Explain My Answer |
application | Duolingo Explain My Answer | Duolingo | Mar 14, 2023 | n/a | limited | custom | GPT-4 API |
model | Claude Instant | Anthropic | Mar 14, 2023 | unknown | limited | unknown | |
model | Claude | Anthropic | Mar 14, 2023 | unknown | limited | unknown | |
model | ChatGLM | ChatGLM | Mar 14, 2023 | 6B parameters (dense) | open | Apache 2.0 | |
application | Anthropic API | Anthropic | Mar 14, 2023 | n/a | limited | none | Claude | Claude Instant |
model | OpenChatKit moderation model | Together | Mar 10, 2023 | 6B parameters (dense) | open | Apache 2.0 | GPT-JT | OIG-moderation |
dataset | OIG-moderation | Together, LAION, Ontocord | Mar 10, 2023 | unknown | open | Apache 2.0 | |
dataset | OIG-43M | Together, LAION, Ontocord | Mar 10, 2023 | 43M instructions | open | Apache 2.0 | P3 | NaturalInstructions-v2 | FLAN dataset |
model | GPT-NeoXT-Chat-Base | Together | Mar 10, 2023 | 20B parameters (dense) | open | Apache 2.0 | GPT-NeoX | OIG-43M |
model | Jurassic-2 | AI21 Labs | Mar 9, 2023 | unknown | limited | unknown | |
application | AI21 Summarization API | AI21 Labs | Mar 9, 2023 | n/a | limited | none | Jurassic-2 |
application | AI21 Paraphrase API | AI21 Labs | Mar 9, 2023 | n/a | limited | none | Jurassic-2 |
model | VisualChatGPT | Microsoft | Mar 8, 2023 | unknown | closed | none | OpenAI API |
application | DuckAssist | DuckDuckGo | Mar 8, 2023 | n/a | open | unknown | Anthropic API |
application | EinsteinGPT | Salesforce | Mar 7, 2023 | n/a | limited | unknown | ChatGPT API |
application | ChatGPT for Slack | OpenAI, Salesforce | Mar 7, 2023 | n/a | limited | unknown | ChatGPT API |
application | Brex Chat | Brex | Mar 7, 2023 | n/a | limited | custom | ChatGPT API |
application | Azure Cognitive Services for Vision | Microsoft | Mar 7, 2023 | n/a | limited | custom | Florence |
model | USM | Mar 6, 2023 | 2B parameters (dense) | limited | unknown | YT-NLU-U | Pub-U | Web-NTL | YT-SUP+ | Pub-S | |
model | PaLM-E | Mar 6, 2023 | 562B parameters (dense) | closed | unknown | PaLM | ViT-22B | |
model | Flan-UL2 | Mar 2, 2023 | 20B parameters (dense) | open | Apache 2.0 | UL2 | Flan Collection | |
dataset | gpt-3.5-turbo dataset | OpenAI | Mar 1, 2023 | unknown | limited | unknown | |
model | gpt-3.5-turbo | OpenAI | Mar 1, 2023 | unknown | limited | custom | gpt-3.5-turbo dataset |
application | Whisper API | OpenAI | Mar 1, 2023 | n/a | open | custom | Whisper |
application | Speak | Speak | Mar 1, 2023 | n/a | open | Whisper API | |
application | Shop Assistant | Shop | Mar 1, 2023 | n/a | open | ChatGPT API | |
application | Q-Chat | Quizlet | Mar 1, 2023 | n/a | open | none | ChatGPT API |
application | My AI for Snapchat | Snap | Mar 1, 2023 | n/a | open | custom | ChatGPT API |
model | KOSMOS-1 | Microsoft | Mar 1, 2023 | 1.6B parameters (dense) | closed | MIT | The Pile | CommonCrawl | LAION-2B-en | LAION-400M | COYO-700M | Conceptual Captions |
application | ChatGPT API | OpenAI | Mar 1, 2023 | n/a | open | custom | ChatGPT |
application | Ask Instacart | Instacart | Mar 1, 2023 | n/a | limited | ChatGPT API | |
model | Vid2Seq | Feb 27, 2023 | 500M parameters (dense) | open | Apache 2.0 | T5 | CLIP | YT-Temporal-1B | |
model | SantaCoder | BigCode | Feb 24, 2023 | 1.1B parameters (dense) | open | Apache 2.0 | The Stack | BigCode Dataset |
model | LLaMA | Meta | Feb 24, 2023 | 65B parameters (dense) | open | LLaMa License (model weights), GPLv3 (code) | CommonCrawl | C4 | Github | Wikipedia | BooksCorpus | arXiv | StackExchange |
application | AI DJ | Spotify | Feb 23, 2023 | n/a | limited | custom | ChatGPT API | Sonantic AI |
application | Notion AI | Notion | Feb 22, 2023 | n/a | limited | Anthropic API | |
application | Cohere Summarize Endpoint | Cohere | Feb 22, 2023 | n/a | limited | Limited use license to Cohere platform users [Terms of Use]. | |
application | Bain Chat | Bain | Feb 21, 2023 | n/a | limited | unknown | ChatGPT API |
dataset | LAION-1B | Alibaba | Feb 20, 2023 | 1B image-text pairs | closed | unknown | LAION-5B |
model | Composer | Alibaba | Feb 20, 2023 | 4.4B parameters (dense) | closed | unknown | ImageNet | WebVision | LAION-1B |
model | ViT-22B | Feb 10, 2023 | 22B parameters (dense) | closed | unknown | JFT | |
dataset | Rater-SF | Feb 8, 2023 | 24k captions | closed | unknown | MusicCaps | |
dataset | Rater-LF | Feb 8, 2023 | 10k captions | closed | unknown | MusicCaps | |
model | Noise2Music pseudolabeler | Feb 8, 2023 | unknown | closed | unknown | MuLan | MuLaMCap | LaMDA-LF | Rater-LF | Rater-SF | |
dataset | Noise2Music pseudolabel dataset | Feb 8, 2023 | 340k hours audio with pseudolabels | closed | unknown | Noise2Music audio dataset | Noise2Music pseudolabeler | |
dataset | Noise2Music audio dataset | Feb 8, 2023 | 340k hours audio | closed | unknown | ||
model | Noise2Music | Feb 8, 2023 | unknown | closed | unknkown | Noise2Music pseudolabel dataset | |
dataset | LaMDA-LF | Feb 8, 2023 | 150k songs | closed | unknown | LaMDA | |
model | Prometheus | Microsoft | Feb 7, 2023 | unknown | closed | unknown | |
application | Bing Search | Microsoft | Feb 7, 2023 | n/a | limited | custom | ChatGPT API |
application | Bard | Feb 6, 2023 | n/a | closed | unknown | LaMDA | |
application | Sage API | OpenAI | Feb 3, 2023 | n/a | limited | unknown | Sage |
model | Sage | OpenAI | Feb 3, 2023 | unknown | limited | unknown | |
application | Poe | Quora | Feb 3, 2023 | n/a | limited | none | ChatGPT API | GPT-4 API | Claude API | Dragonfly API | Sage API |
application | Dragonfly API | OpenAI | Feb 3, 2023 | n/a | limited | unknown | Dragonfly |
model | Dragonfly | OpenAI | Feb 3, 2023 | unknown | limited | unknown | |
application | UnderwriteGPT | Paladin Group and Dais Technology | Feb 1, 2023 | n/a | limited | ||
dataset | Phenaki Video-Text Corpus | Feb 1, 2023 | 15M text-video pairs at 8FPS | closed | unknown | ||
model | Phenaki | Feb 1, 2023 | 1.8B parameters (dense) | closed | unknown | LAION-400M | Phenaki Video-Text Corpus | |
dataset | Flan Collection | Jan 31, 2023 | 1836 tasks | open | Apache 2.0 | Flan dataset | P3 | NaturalInstructions-v2 | |
application | ChatGPT powered by OBO | HubSpot | Jan 31, 2023 | n/a | limited | unknown | ChatGPT API |
model | w2v-BERT | Jan 26, 2023 | 600M parameters (dense) | closed | unknown | Free Music Archive | |
model | SoundStream | Jan 26, 2023 | unknown | closed | unknown | Free Music Archive | |
model | MusicLM semantic model | Jan 26, 2023 | 430M parameters (dense) | closed | unknown | MusicLM dataset | |
dataset | MusicLM dataset | Jan 26, 2023 | 280K hours audio | closed | unknown | ||
model | MusicLM acoustic model | Jan 26, 2023 | 430M parameters (dense) | closed | unknown | MusicLM dataset | |
model | MusicLM | Jan 26, 2023 | 1.4B parameters (dense) | closed | unknown | SoundStream | w2v-BERT | MuLan | MusicLM semantic model | MusicLM acoustic model | |
dataset | OpenAI toxicity dataset | OpenAI | Jan 18, 2023 | unknown | closed | unknown | |
model | OpenAI toxicity classifier | OpenAI | Jan 18, 2023 | unknown | closed | unknown | OpenAI toxicity dataset |
application | NeevaAI | Neeva | Jan 6, 2023 | n/a | open | Custom | Neeva model |
model | VALL-E | Microsoft | Jan 5, 2023 | unknown | closed | unknown | |
model | Palmyra | Writer | Jan 1, 2023 | 20B parameters (dense) | open | Apache 2.0 | Writer dataset |
model | Cohere Command | Cohere | Jan 1, 2023 | unknown | limited | unknown | Cohere Base |
model | MultiMedQA | Dec 26, 2022 | unknown | closed | unknown | MedQA | MedMCQA | PubMedQA | MMLU | LiveQA | Medication QA | HealthSearchQA | |
model | Med-PaLM | Dec 26, 2022 | 540B parameters (dense) | closed | unknown | Flan-PaLM | MultiMedQA | |
model | OPT-IML | Meta | Dec 22, 2022 | 175B parameters (dense) | open | OPT-IML 175B License | OPT | OPT-IML Bench |
application | Bird SQL | Perplexity | Dec 15, 2022 | n/a | closed | none | Perplexity Ask | OpenAI API |
model | BioMedLM | Stanford | Dec 15, 2022 | 2.7B parameters (dense) | open | bigscience-bloom-rail-1.0 | The Pile |
dataset | LAION-5B | LAION | Dec 12, 2022 | 5B image-text pairs | open | CC BY 4.0 | CLIP | mCLIP | CommonCrawl |
dataset | LAION-2B-en | LAION | Dec 12, 2022 | 2.32B image-text pairs | open | CC BY 4.0 | CLIP | LAION-5B |
model | Cohere Embed (Multilingual) | Cohere | Dec 12, 2022 | unknown | limited | unknown | |
application | Perplexity Ask | Perplexity | Dec 7, 2022 | n/a | open | none | GPT-3.5 | Bing Search |
model | InternVideo | Shanghai AI Laboratory | Dec 6, 2022 | 1.3B parameters (dense) | open | Apache 2.0 | Kinetics-400 | WebVid-2M | WebVid-10M | HowTo100M | AVA | Something-Something-v2 | Kinetics-710 |
dataset | Jurassic-1 Instruct dataset | AI21 Labs | Dec 1, 2022 | unknown | closed | unknown | |
model | Jurassic-1 Instruct | AI21 Labs | Dec 1, 2022 | 17B parameters (dense) | limited | unknown | Jurassic-1 | Jurassic-1 Instruct dataset |
model | text-davinci-003 | OpenAI | Nov 30, 2022 | unknown | limited | unknown | text-davinci-002 |
application | ChatGPT | OpenAI | Nov 30, 2022 | n/a | open | custom | gpt-3.5-turbo | OpenAI toxicity classifier |
model | GPT-JT | Together | Nov 29, 2022 | 6B parameters (dense) | open | Apache 2.0 | GPT-J | P3 | NaturalInstructions-v2 |
model | RoentGen | Stanford | Nov 23, 2022 | 330M parameters (dense) | open | Stable Diffusion | RoentGen radiology dataset | |
model | Florence | Microsoft | Nov 23, 2022 | 900M parameters (dense) | closed | unknown | FLD-900M |
dataset | FLD-900M | Microsoft | Nov 23, 2022 | 900M image-text pairs | closed | unknown | |
dataset | The Stack | BigCode | Nov 20, 2022 | 3.1 TB | open | Apache 2.0 | GitHub |
model | OpenFold | Columbia | Nov 20, 2022 | open | CC BY 4.0 | AlphaFold2 | OpenProteinSet | |
dataset | The Galactica Corpus | Meta | Nov 15, 2022 | 106B tokens | closed | unknown | CommonCrawl | Wikipedia | arXiv |
model | Galactica | Meta | Nov 15, 2022 | 120B parameters (dense) | open | CC BY-NC 4.0 | The Galactica Corpus |
dataset | xP3 | BigScience | Nov 3, 2022 | 9.4GB | open | Apache 2.0 | P3 |
model | BLOOMZ | BigScience | Nov 3, 2022 | 176B parameters (dense) | open | BigScience RAIL v1.0 | BLOOM | xP3 |
model | ESM-2 | Meta | Oct 31, 2022 | 15B parameters (dense) | open | MIT | UniRef50 | UniRef90 |
model | ERNIE-ViLG 2.0 | Baidu | Oct 27, 2022 | 10B parameters (dense) | closed | unknown | |
model | MAGMA | Aleph Alpha | Oct 24, 2022 | 6B parameters (dense) | open | MIT | GPT-J | CLIP |
model | U-PaLM | Oct 20, 2022 | 540B parameters (dense) | closed | unknown | PaLM | PaLM dataset | |
model | Flan-U-PaLM | Oct 20, 2022 | 540B parameters (dense) | closed | unknown | U-PaLM | Muffin | P3 | NaturalInstructions-v2 | |
model | Flan-T5 | Oct 20, 2022 | 11B parameters (dense) | open | Apache 2.0 | T5 | Muffin | P3 | NaturalInstructions-v2 | Flan CoT | |
model | Flan-PaLM | Oct 20, 2022 | 540B parameters (dense) | closed | unknown | PaLM | Muffin | P3 | NaturalInstructions-v2 | |
dataset | P3 | BigScience | Oct 15, 2022 | 2000 prompts | open | Apache 2.0 | |
model | GenSLM | Argonne National Laboratory | Oct 11, 2022 | 25B parameters (dense) | open | MIT | SARS-CoV-2 genome dataset | BV-BRC dataset |
dataset | VIMA dataset | NVIDIA, Stanford | Oct 6, 2022 | 200M parameters (dense model) | open | MIT | T5 | Mask R-CNN | VIMA dataset |
model | VIMA | NVIDIA, Stanford | Oct 6, 2022 | 200M parameters (dense) | open | MIT | |
dataset | Make-A-Video dataset | Meta | Sep 29, 2022 | 20M video clips, 2.3B image-text pairs | limited | none | LAION-5B | WebVid-10M | HD-VILA-100M |
model | Make-A-Video | Meta | Sep 29, 2022 | unknown | closed | none | Make-A-Video dataset |
model | Dramatron | DeepMind | Sep 29, 2022 | 70B parameters (dense) | closed | unknown | Chinchilla |
model | T-ULRv5 | Microsoft | Sep 28, 2022 | 2.2B parameters (dense) | limited | unknown | |
dataset | Sparrow response preference dataset | DeepMind | Sep 28, 2022 | 72k comparisons | closed | unknown | Chinchilla |
dataset | Sparrow adversarial probing dataset | DeepMind | Sep 28, 2022 | 27k ratings | closed | unknown | Chinchilla |
model | Sparrow Rule reward model | DeepMind | Sep 28, 2022 | 70B parameters (dense) | closed | unknown | Chinchilla | Sparrow adversarial probing dataset |
model | Sparrow Preference reward model | DeepMind | Sep 28, 2022 | 70B parameters (dense) | closed | unknown | Chinchilla | Sparrow response preference dataset |
model | Sparrow | DeepMind | Sep 28, 2022 | 70B parameters (dense) | closed | unknown | Chinchilla | Google Search | Sparrow Rule reward model | Sparrow Preference reward model |
model | BioGPT | Microsoft | Sep 24, 2022 | 1.5B parameters (dense) | open | MIT | PubMed |
dataset | Whisper dataset | OpenAI | Sep 21, 2022 | 680k hours | closed | unknown | |
model | Whisper | OpenAI | Sep 21, 2022 | 1.5B parameters (dense) | open | MIT | Whisper dataset |
model | CodeGeeX | Tsinghua | Sep 20, 2022 | 13B parameters (dense) | limited | Apache 2.0 | |
dataset | WebLI | Sep 14, 2022 | 10B images, 12B alt-text | closed | unknown | ||
model | ViT-e | Sep 14, 2022 | 3.9B parameters (dense) | closed | unknown | JFT | |
model | PaLI | Sep 14, 2022 | 17B parameters (dense) | closed | unknown | mT5 | ViT-e | WebLI | |
model | ACT-1 | Adept | Sep 14, 2022 | closed | unknown | ||
model | AudioLM | Sep 7, 2022 | 1B parameters (dense) | closed | unknown | w2v-BERT | SoundStream | |
model | VQGAN-CLIP | EleutherAI | Sep 4, 2022 | 227M parameters (dense) | open | MIT | VQGAN | CLIP |
dataset | COYO-700M | Kakao Brain | Aug 31, 2022 | 747M image-text pairs | open | CC-BY-4.0 | CommonCrawl |
model | BEiT-3 | Microsoft | Aug 31, 2022 | 1.9B parameters (dense) | open | Multiway Transformer network | |
dataset | MuLan dataset | Aug 26, 2022 | 370K hours audio | closed | unknown | ||
model | MuLan | Aug 26, 2022 | unknown | closed | unknown | AST | BERT | MuLan dataset | |
application | AI Test Kitchen | Aug 25, 2022 | n/a | limited | unknown | LaMDA | |
model | PEER | Meta | Aug 24, 2022 | 3B parameters (dense) | open | ||
application | Stable Diffusion | Stability AI | Aug 22, 2022 | n/a | open | custom | |
model | PaLM-SayCan | Aug 16, 2022 | 540B parameters (dense) | closed | unknown (model weights), Apache 2.0 (SayCan code) | PaLM | |
application | OpenAI Moderation API | OpenAI | Aug 10, 2022 | n/a | open | custom | OpenAI toxicity classifier |
model | GLM-130B | Tsinghua | Aug 4, 2022 | 130B parameters (dense) | open | GLM-130B License | The Pile | GLM-130B Chinese corpora | P3 | DeepStruct finetuning dataset |
model | BLOOM | BigScience | Jul 12, 2022 | 176B parameters (dense) | open | BigScience RAIL v1.0 | ROOTS |
dataset | Minerva Math Web Pages dataset | Jun 29, 2022 | 17.5B tokens | closed | unknown | ||
model | Minerva | Jun 29, 2022 | 540B parameters (dense) | closed | unknown | PaLM | arXiv | PaLM dataset | Minerva Math Web Pages dataset | |
dataset | web_clean | OpenAI | Jun 23, 2022 | 70k hours | closed | unknown | |
application | Yandex Search | Yandex | Jun 23, 2022 | n/a | open | custom | YaLM |
model | VPT | OpenAI | Jun 23, 2022 | 500M parameters (dense) | open | MIT | web_clean |
model | YaLM | Yandex | Jun 22, 2022 | 100B parameters (dense) | open | Apache 2.0 | The Pile | Yandex Russian Pretraining Dataset |
model | Parti | Jun 22, 2022 | 20B parameters (dense) | closed | unknown | C4 | LAION-400M | FIT400M | JFT-4B | |
dataset | MineDojo | NVIDIA | Jun 17, 2022 | 730k videos, 6k Wikipedia pages, 340k reddit posts | open | MIT | YouTube | Wikipedia | Reddit |
dataset | ROOTS | BigScience | Jun 6, 2022 | 1.6TB | open | custom | |
model | CogVideo | Tsinghua | May 29, 2022 | unknown | open | Apache 2.0 | |
model | Imagen | May 23, 2022 | 14B parameters (dense) | open | unknown | LAION-400M | Google internal image-text dataset | |
dataset | Gato dataset | DeepMind | May 12, 2022 | 10.5 TB Text, 2.2B Text-Image pairs, 1.5T tokens of simulated control, 500k robotics trajectories | closed | unknown | MassiveText |
model | Gato | DeepMind | May 12, 2022 | 1.2B parameters (dense) | closed | unknown | Gato dataset |
model | UL2 | May 10, 2022 | 20B parameters (dense) | open | Apache 2.0 | C4 | |
application | Cohere Classify Endpoint | Cohere | May 5, 2022 | n/a | limited | Limited use license to Cohere platform users [Terms of Use]. | Cohere Embed (Multilingual) | Cohere Embed (English) |
model | text-davinci-002 | OpenAI | May 1, 2022 | unknown | limited | unknown | code-davinci-002 |
dataset | code-davinci-002 dataset | OpenAI | May 1, 2022 | unknown | limited | unknown | |
model | code-davinci-002 | OpenAI | May 1, 2022 | unknown | limited | unknown | code-davinci-002 dataset |
model | OPT | Meta | May 1, 2022 | 175B parameters (dense) | limited | OPT-175B License | RoBERTa dataset | The Pile | PushShift.io Reddit |
dataset | M3W | DeepMind | Apr 29, 2022 | 182GB Text, 185M Images | closed | unknown | |
model | Flamingo | DeepMind | Apr 29, 2022 | 80B parameters (dense) | closed | unknown | M3W | ALIGN | LTIP | VTP | Chinchilla |
model | CogView 2 | Tsinghua | Apr 28, 2022 | 6B parameters (dense) | open | Apache 2.0 | |
model | VATT | Apr 22, 2022 | 155M parameters (dense) | open | Apache 2.0 | AudioSet | HowTo100M | |
dataset | RedPajama-Data | Together | Apr 17, 2022 | 1.2 trillion tokens | open | Apache 2.0 | GitHub | Wikipedia |
dataset | NaturalInstructions-v2 | AI2 | Apr 16, 2022 | 1600 tasks | open | Apache 2.0 | |
dataset | Luminous dataset | Aleph Alpha | Apr 14, 2022 | unknown | closed | unknown | |
model | Luminous | Aleph Alpha | Apr 14, 2022 | 200B parameters (dense) | limited | none | Luminous dataset |
model | DALL·E 2 | OpenAI | Apr 13, 2022 | unknown | limited | unknown | DALL·E dataset | CLIP dataset |
model | InCoder | Meta, CMU, TTI-Chicago, UC Berkeley, University of Washington | Apr 12, 2022 | 6B parameters (dense) | open | CC BY-NC 4.0 | |
model | Anthropic RLHF models | Anthropic | Apr 12, 2022 | 52B parameters (dense) | closed | Anthropic Harmlessness dataset | Anthropic Helpfulness dataset | |
application | Anthropic Human Feedback Interface | Anthropic | Apr 12, 2022 | n/a | closed | unknown | Anthropic RLHF models |
dataset | Anthropic Helpfulness dataset | Anthropic | Apr 12, 2022 | 271.5 MB | open | MIT | Anthropic Human Feedback Interface |
dataset | Anthropic Harmlessness dataset | Anthropic | Apr 12, 2022 | unknown | closed | unknown | Anthropic Human Feedback Interface |
dataset | PaLM dataset | Apr 4, 2022 | 3.92 TB | closed | unknown | Infiniset | |
model | PaLM | Apr 4, 2022 | 540B parameters (dense) | limited | unknown | PaLM dataset | |
model | Chinchilla | DeepMind | Mar 29, 2022 | 70B parameters (dense) | closed | unknown | MassiveText |
model | CodeGen | Salesforce | Mar 25, 2022 | 16B parameters (dense) | open | none (model weights), BSD-3-Clause (code) | |
model | GopherCite reward model | DeepMind | Mar 16, 2022 | 7B parameters (dense) | closed | unknown | Gopher | GopherCite Preference dataset |
dataset | GopherCite Preference dataset | DeepMind | Mar 16, 2022 | 33k response pairs | closed | unknown | Gopher | Google Search |
model | GopherCite | DeepMind | Mar 16, 2022 | 280B parameters (dense) | closed | unknown | Gopher | Google Search | GopherCite reward model |
model | PolyCoder | CMU | Feb 26, 2022 | 2.7B parameters (dense) | open | MIT | Github |
model | GPT-NeoX | EleutherAI | Feb 2, 2022 | 20B parameters (dense) | open | Apache 2.0 | The Pile |
model | AlphaCode | DeepMind | Feb 2, 2022 | 41B parameters (dense) | closed | unknown | |
model | Megatron-Turing NLG | Microsoft, NVIDIA | Jan 28, 2022 | 530B parameters (dense) | limited | unknown | The Pile |
dataset | LAION-115M | Salesforce | Jan 28, 2022 | 115M image-text pairs | open | BSD-3-Clause | LAION-400M |
model | BLIP | Salesforce | Jan 28, 2022 | unknown | open | BSD-3-Clause | ViT-B | BERT | COCO | Visual Genome | Conceptual Captions | Conceptual 12M | SBU Captions | LAION-115M |
model | InstructGPT | OpenAI | Jan 27, 2022 | 175B parameters (dense) | closed | unknown | GPT-3 | OpenAI API |
dataset | YT-Temporal-1B | University of Washington | Jan 7, 2022 | 20M videos | open | MIT | YouTube |
application | AssemblyAI | AssemblyAI | 2022 | n/a | limited | custom | Anthropic API |
model | ERNIE-ViLG | Baidu | Dec 31, 2021 | 10B parameters (dense) | limited | none | |
model | ERNIE 3.0 Titan | Baidu, PengCheng Laboratory | Dec 23, 2021 | 260B parameters (dense) | closed | unknown | |
dataset | GLaM Web dataset | Dec 13, 2021 | unknown | closed | unknown | ||
dataset | GLaM News dataset | Dec 13, 2021 | unknown | closed | unknown | ||
dataset | GLaM Forums dataset | Dec 13, 2021 | unknown | closed | unknown | ||
dataset | GLaM Conversations dataset | Dec 13, 2021 | unknown | closed | unknown | ||
model | GLaM | Dec 13, 2021 | 1.2T parameters (sparse) | closed | unknown | GLaM Web dataset | Wikipedia | GLaM Conversations dataset | GLaM Forums dataset | BooksCorpus | GLaM News dataset | |
model | RETRO | DeepMind | Dec 8, 2021 | 7.5B parameters (dense) | closed | unknown | MassiveText |
dataset | PMD | Meta | Dec 8, 2021 | 70M | closed | unknown | COCO | YFCC100M | SBU Captions | Localized Narratives | Visual Genome | Wikipedia | Conceptual Captions | Red Caps |
dataset | MassiveText | DeepMind | Dec 8, 2021 | 10.5 TB | closed | unknown | |
model | Gopher | DeepMind | Dec 8, 2021 | 280B parameters (dense) | closed | unknown | MassiveText |
model | FLAVA | Meta | Dec 8, 2021 | 306M | open | BSD-3-Clause | PMD |
model | CodeParrot | HuggingFace | Dec 6, 2021 | 1B parameters (dense) | open | none | |
model | Turing NLR-v5 | Microsoft | Dec 2, 2021 | 5B parameters (dense) | limited | unknown | |
application | Wordtune Read | AI21 Labs | Nov 16, 2021 | n/a | limited | Wordtune License | AI21 Summarize API |
dataset | coheretext | Cohere | Nov 15, 2021 | 200 GB | closed | unknown | |
application | Cohere Generate Endpoint | Cohere | Nov 15, 2021 | n/a | limited | Limited use license to Cohere platform users [Terms of Use]. | Cohere Base | Cohere Command |
application | Cohere Embed Endpoint | Cohere | Nov 15, 2021 | n/a | limited | Limited use license to Cohere platform users [Terms of Use]. | Cohere Embed (Multilingual) | Cohere Embed (English) |
model | Cohere Embed (English) | Cohere | Nov 15, 2021 | unknown | limited | unknown | |
model | Cohere Base | Cohere | Nov 15, 2021 | unknown | limited | unknown | coheretext |
application | Cohere API | Cohere | Nov 15, 2021 | n/a | limited | custom | Cohere Generate Endpoint | Cohere Embed Endpoint | Cohere Classify Endpoint | Cohere Summarize Endpoint |
model | VLMo | Microsoft | Nov 3, 2021 | 562M parameters (dense) | closed | none | Conceptual Captions | SBU Captions | COCO | Visual Genome | Wikipedia | BooksCorpus |
model | mT0 | BigScience | Oct 15, 2021 | 13B parameters (dense) | open | BigScience RAIL v1.0 | mT5 | xP3 |
model | T0++ | BigScience | Oct 15, 2021 | 11B parameters (dense) | open | Apache 2.0 | T5 | P3 |
application | Aleph Alpha API | Aleph Alpha | Sep 30, 2021 | n/a | limited | none | Luminous |
dataset | Muffin | Sep 3, 2021 | 62 tasks | open | Apache 2.0 | ||
dataset | LAION-400M | LAION | Aug 20, 2021 | 400M image-text pairs | open | CC BY 4.0 | CLIP | CommonCrawl |
dataset | Jurassic-1 dataset | AI21 Labs | Aug 11, 2021 | 300B tokens | closed | unknown | |
model | Jurassic-1 | AI21 Labs | Aug 11, 2021 | 178B parameters (dense) | limited | unknown | Jurassic-1 dataset |
application | AI21 Playground | AI21 Labs | Aug 11, 2021 | n/a | limited | none | Jurassic-1 | Jurassic-1 Instruct | Jurassic-2 | AI21 Summarization API | AI21 Paraphrase API |
dataset | HumanEval | OpenAI | Aug 10, 2021 | 214 KB | open | MIT | |
dataset | Codex dataset | OpenAI | Aug 10, 2021 | 159 GB | closed | ||
model | Codex | OpenAI | Aug 10, 2021 | 12B parameters (dense) | limited | unknown | GPT-3 | Codex dataset | HumanEval |
model | AlphaFold2 | DeepMind | Jul 15, 2021 | 93M parameters (dense) | open | Apache 2.0 | Protein Data Bank |
application | GitHub CoPilot | Microsoft | Jun 29, 2021 | n/a | limited | unknown | Codex |
model | LaMDA | Jun 18, 2021 | 137B parameters (dense) | closed | unknown | Infiniset | |
dataset | Infiniset | Jun 18, 2021 | unknown | closed | unknown | ||
model | GPT-J | EleutherAI | Jun 4, 2021 | 6B parameters (dense) | open | Apache 2.0 | The Pile |
model | CogView | Tsinghua | May 26, 2021 | 4B parameters (dense) | open | Apache 2.0 | |
model | HyperCLOVA | Naver | May 21, 2021 | 82B parameters (dense) | closed | unknown | |
dataset | MUM dataset | May 18, 2021 | unknown | closed | unknown | ||
model | MUM | May 18, 2021 | unknown | closed | unknown | MUM dataset | |
model | Docugami | Microsoft | Apr 12, 2021 | 20B parameters (dense) | limited | ||
model | Megatron-LM | NVIDIA | Apr 9, 2021 | 1T parameters (dense) | closed | unknown | |
dataset | WebVid-2M | University of Oxford | Apr 1, 2021 | 2.5M video-text pairs, 13K hours video | open | WebVid Dataset Terms | WebVid-10M |
dataset | WebVid-10M | University of Oxford | Apr 1, 2021 | 10.7M video-text pairs, 52K hours video | open | WebVid Dataset Terms | |
application | Crisis Contact Simulator | The Trevor Project | Mar 24, 2021 | n/a | closed | unknown | OpenAI API |
model | GPT-Neo | EleutherAI | Mar 21, 2021 | 2.7B parameters (dense) | open | MIT | The Pile |
dataset | Conceptual 12M | Feb 17, 2021 | 12M (image, text) pairs | open | Conceptual Captions License | ||
dataset | Wu Dao dataset | Beijing Academy of Artificial Intelligence | Jan 12, 2021 | unknown | closed | unknown | |
model | Wu Dao 2.0 | Beijing Academy of Artificial Intelligence | Jan 12, 2021 | 1.75T parameters (dense) | closed | unknown | Wu Dao dataset |
dataset | DALL·E dataset | OpenAI | Jan 5, 2021 | 250M (image, text) pairs | closed | unknown | |
model | DALL·E | OpenAI | Jan 5, 2021 | 12B parameters (dense) | limited | unknown | DALL·E dataset |
dataset | CLIP dataset | OpenAI | Jan 5, 2021 | 400M (image, text) pairs | closed | unknown | |
model | CLIP | OpenAI | Jan 5, 2021 | unknown | open | MIT | CLIP dataset |
dataset | The Pile | EleutherAI | Jan 1, 2021 | 825 GB | open | MIT | |
application | Wordtune | AI21 Labs | Oct 27, 2020 | n/a | limited | Wordtune License | AI21 Paraphrase API |
application | OpenAI API | OpenAI | Jun 11, 2020 | n/a | limited | custom | GPT-3 | Codex | code-davinci-002 | text-davinci-002 | text-davinci-003 | gpt-3.5-turbo | Whisper | DALL·E | GPT-4 |
dataset | GPT-3 dataset | OpenAI | Jun 11, 2020 | 570 GB | closed | unknown | WebText |
model | GPT-3 | OpenAI | Jun 11, 2020 | 175B parameters (dense) | limited | unknown | GPT-3 dataset |
model | Jukebox | OpenAI | Apr 30, 2020 | 5B parameters (dense) | open | Noncommercial Use License | Jukebox Dataset |
application | AI Dungeon | Latitude | Dec 17, 2019 | n/a | limited | custom | OpenAI API |
dataset | Internal Google BERT dataset | Nov 25, 2019 | unknown | closed | unknown | ||
model | Internal Google BERT | Nov 25, 2019 | unknown | closed | unknown | Internal Google BERT dataset | |
application | Google Search | Nov 25, 2019 | n/a | open | none | Internal Google BERT | MUM | |
dataset | WebText | OpenAI | Nov 1, 2019 | 40 GB | closed | unknown | |
model | GPT-2 | OpenAI | Nov 1, 2019 | 1.5B parameters (dense) | open | Modified MIT License | WebText |
model | T5 | Oct 23, 2019 | 11B parameters (dense) | open | Apache 2.0 | C4 | |
dataset | C4 | Oct 23, 2019 | 750GB | open | ODC-By 1.0 | CommonCrawl | |
model | UniLM | Microsoft | Oct 1, 2019 | 340M parameters (dense) | open | MIT | |
dataset | HowTo100M | École Normale Supérieure, Inria | Jun 7, 2019 | 136M video clips | open | Apache 2.0 | YouTube |
dataset | Conceptual Captions | Jul 1, 2018 | 3.3M (image, text) pairs | open | Conceptual Captions License | ||
dataset | SBU Captions | Stony Brook University | Dec 12, 2011 | 1M image-text pairs | open | none | Flickr |
application | YouTube | Feb 14, 2005 | n/a | open | USM | ||
model | You model | You | unknown | unknkown | closed | unknown | You dataset |
dataset | You dataset | You | unknown | unknown | closed | unknown | |
application | You Search | You | unknown | n/a | open | unknown | You model |
application | Viable | Viable | unknown | n/a | limited | unknown | OpenAI API |
application | Sana | Sana | unknown | n/a | limited | custom | OpenAI API |
application | Robin AI | Robin AI | unknown | n/a | limited | none | Anthropic API |
model | Neeva model | Neeva | unknown | unknown | closed | unknown | Neeva dataset |
dataset | Neeva dataset | Neeva | unknown | unknown | closed | unknown | |
application | Microsoft Word | Microsoft | unknown | n/a | open | custom | Microsoft 365 Copilot |
application | Microsoft Teams | Microsoft | unknown | n/a | open | custom | Microsoft 365 Copilot | Microsoft Business Chat |
application | Microsoft Suggested Replies | Microsoft | unknown | n/a | limited | custom | |
application | Microsoft PowerPoint | Microsoft | unknown | n/a | open | custom | Microsoft 365 Copilot |
application | Microsoft Power Platform | Microsoft | unknown | n/a | limited | custom | Microsoft 365 Copilot |
application | Microsoft Outlook | Microsoft | unknown | n/a | open | custom | Microsoft 365 Copilot |
application | Microsoft Inside Look | Microsoft | unknown | n/a | limited | custom | |
application | Microsoft Excel | Microsoft | unknown | n/a | open | custom | Microsoft 365 Copilot |
application | unknown | n/a | open | unknown | Azure Cognitive Services for Vision | ||
application | Juni Tutor Bot | Juni Learning | unknown | n/a | limited | unknown | Anthropic API |
application | HyperWrite | OthersideAI | unknown | n/a | limited | custom | OpenAI API |
application | GooseAI API | GooseAI | unknown | n/a | limited | custom | GPT-NeoX |