top 10 everything
suggest a leaderboardtop 10 open-source LLMs (based on huggingface downloads or github forks)
| description | huggingface | github | paper | license | ||
|---|---|---|---|---|---|---|
1 | bert base uncased | a transformers model pretrained on a large corpus of English data in a self-supervised fashion. | https://huggingface.co/bert-base-uncased | https://github.com/google-research/bert | https://arxiv.org/abs/1810.04805 | Apache 2.0 |
2 | Dolly | an instruction-following large language model | https://huggingface.co/databricks/dolly-v2-12b | https://github.com/databrickslabs/dolly#getting-started-with-response-generation | https://www.databricks.com/blog/2023/04/12/dolly-first-open-commercially-viable-instruction-tuned-llm | MIT |
3 | bloom | an autoregressive Large Language Model | https://huggingface.co/bigscience/bloom | https://github.com/bigscience-workshop/bigscience | https://arxiv.org/abs/2211.05100 | OpenRAIL-M v1 |
4 | Cerebras-GPT-13b | released to facilitate research into LLM scaling laws | https://huggingface.co/cerebras/Cerebras-GPT-13B | https://github.com/Cerebras | https://arxiv.org/abs/2304.03208 | Apache 2.0 |
5 | mpt-7b | MPT-7B is a decoder-style transformer pretrained from scratch on 1T tokens of English text and code | https://huggingface.co/mosaicml/mpt-7b | https://github.com/mosaicml/llm-foundry | https://www.mosaicml.com/blog/mpt-7b | Apache 2.0, CC BY-SA-3.0 |
6 | StableLM-Alpha | StableLM-Alpha models are trained on the new dataset that build on The Pile | https://huggingface.co/stabilityai/stablelm-base-alpha-3b | https://github.com/Stability-AI/StableLM#stablelm-alpha | https://stability.ai/blog/stability-ai-launches-the-first-of-its-stablelm-suite-of-language-models | CC BY-SA-4.0 |
7 | replit-code-v1-3b | a 2.7B Causal Language Model focused on Code Completion | https://huggingface.co/replit/replit-code-v1-3b | https://github.com/replit/ReplitLM | none | CC BY-SA-4.0 |
8 | starcoder | 15.5B parameter models trained on 80+ programming languages | https://huggingface.co/bigcode/starcoder#%23model-summary | https://github.com/bigcode-project/starcoder | https://drive.google.com/file/d/1cN-b9GnWtHzQRoE7M7gAEyivY0kl4BYs/view | OpenRAIL-M v1 |
9 | h2oGPT | the worlds best open source GPT | none | https://github.com/h2oai/h2ogpt | https://github.com/h2oai/h2ogpt | Apache 2.0 |
10 | open llama | an open source reproduction of Meta AIs LLaMA large language model | https://huggingface.co/openlm-research/open_llama_7b_preview_200bt | https://github.com/openml/OpenML | https://arxiv.org/abs/1407.7722 | Apache 2.0 |
powered by flywheel