{"id":19722066,"url":"https://github.com/tensor-fusion/sophia-jax","last_synced_at":"2025-07-14T22:32:34.922Z","repository":{"id":245462427,"uuid":"804943384","full_name":"tensor-fusion/sophia-jax","owner":"tensor-fusion","description":"JAX implementation of 'Sophia: A Scalable Stochastic Second-order Optimizer for Language Model Pre-training'","archived":false,"fork":false,"pushed_at":"2024-05-23T21:53:01.000Z","size":265,"stargazers_count":2,"open_issues_count":0,"forks_count":0,"subscribers_count":1,"default_branch":"main","last_synced_at":"2025-02-28T00:33:03.883Z","etag":null,"topics":["deep-learning","jax","large-language-models","llm","machine-learning","optimization","optimizers","sophia"],"latest_commit_sha":null,"homepage":"","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":null,"status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/tensor-fusion.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":null,"code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null}},"created_at":"2024-05-23T15:14:09.000Z","updated_at":"2024-11-06T23:05:54.000Z","dependencies_parsed_at":"2024-06-22T08:09:32.147Z","dependency_job_id":"18c1290f-e945-404d-9a18-543fea176d06","html_url":"https://github.com/tensor-fusion/sophia-jax","commit_stats":null,"previous_names":["tensor-fusion/sophia-jax"],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/tensor-fusion/sophia-jax","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/tensor-fusion%2Fsophia-jax","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/tensor-fusion%2Fsophia-jax/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/tensor-fusion%2Fsophia-jax/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/tensor-fusion%2Fsophia-jax/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/tensor-fusion","download_url":"https://codeload.github.com/tensor-fusion/sophia-jax/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/tensor-fusion%2Fsophia-jax/sbom","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":265360311,"owners_count":23752678,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["deep-learning","jax","large-language-models","llm","machine-learning","optimization","optimizers","sophia"],"created_at":"2024-11-11T23:16:21.399Z","updated_at":"2025-07-14T22:32:34.867Z","avatar_url":"https://github.com/tensor-fusion.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Sophia - JAX\n\n\u003cimg src=\"./sophia.png\" width=\"700px\"\u003e\u003c/img\u003e\n\nJAX implementation of the [Sophia optimizer](https://arxiv.org/abs/2305.14342) for LLM pre-training. Official PyTorch implementation is here: https://github.com/Liuhong99/Sophia\n\nIn the paper, Sophia is reported to be 2x faster than Adam on GPT-2.\n\nIn the wild it's recently been battle-tested on large-scale runs at Meta and a similar speed-up was observed as well: https://x.com/ArmenAgha/status/1780149168692158658\n\n\n## TODO\n- [ ] Reproduce pretraining results with GPT models\n- [ ] Comparisons to AdamW, LION, etc.\n- [ ] etc\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Ftensor-fusion%2Fsophia-jax","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Ftensor-fusion%2Fsophia-jax","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Ftensor-fusion%2Fsophia-jax/lists"}