{"payload":{"feedbackUrl":"https://github.com/orgs/community/discussions/53140","repo":{"id":705054246,"defaultBranch":"main","name":"web-llm","ownerLogin":"Navezjt","currentUserCanPush":false,"isFork":true,"isEmpty":false,"createdAt":"2023-10-14T22:48:20.000Z","ownerAvatar":"https://avatars.githubusercontent.com/u/110367357?v=4","public":true,"private":false,"isOrgOwned":false},"refInfo":{"name":"","listCacheKey":"v0:1697323702.299273","currentOid":""},"activityList":{"items":[{"before":"0d5ac9e6512fb518f08312b90d3afa3c1be5102d","after":"7578f2a50bcea8e9ce0c5d8a72a3ab5712d8e82c","ref":"refs/heads/main","pushedAt":"2023-12-20T16:30:02.000Z","pushType":"push","commitsCount":17,"pusher":{"login":"Navezjt","name":"Joel Navez CT","path":"/Navezjt","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/110367357?s=80&v=4"},"commit":{"message":"[Version] Bump version to 0.2.14 (#245)\n\nBump version to 0.2.14. Main changes include:\r\n- Fix wizard math conversation template\r\n- Change the way we map model weights and model library (deprecate model\r\nlib map)\r\n- FIx Mistral models template\r\n- Fix tokenizer issue (especially stop token)","shortMessageHtmlLink":"[Version] Bump version to 0.2.14 (<a class=\"issue-link js-issue-link\" data-error-text=\"Failed to load title\" data-id=\"2049570178\" data-permission-text=\"Title is private\" data-url=\"https://github.com/mlc-ai/web-llm/issues/245\" data-hovercard-type=\"pull_request\" data-hovercard-url=\"/mlc-ai/web-llm/pull/245/hovercard\" href=\"https://github.com/mlc-ai/web-llm/pull/245\">mlc-ai#245</a>)"}},{"before":"72f1e42ed6f78adf6d63c15b937a020e07c76759","after":"0d5ac9e6512fb518f08312b90d3afa3c1be5102d","ref":"refs/heads/main","pushedAt":"2023-11-27T23:14:06.000Z","pushType":"push","commitsCount":3,"pusher":{"login":"Navezjt","name":"Joel Navez CT","path":"/Navezjt","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/110367357?s=80&v=4"},"commit":{"message":"Add 128MB buffer size friendly models (#213)\n\nThis PR adds three models with 1k context length that can be run by\r\ndevices with `maxStorageBufferBindingSize = 128MB`:\r\n- `Llama-2-7b-chat-hf-q4f16_1-1k`\r\n- `RedPajama-INCITE-Chat-3B-v1-q4f16_1-1k`\r\n- `RedPajama-INCITE-Chat-3B-v1-q4f32_1-1k`\r\n\r\nThis is enabled by https://github.com/apache/tvm/pull/16108, and offers\r\na temporary solution for https://github.com/mlc-ai/web-llm/issues/209\r\nand the discussion in https://github.com/apache/tvm/pull/14640.\r\n\r\nIrrelevantly, we also update RedPajama models from `_0` quantization to\r\n`_1`.\r\n\r\ncc: @tqchen @beaufortfrancois @DustinBrett","shortMessageHtmlLink":"Add 128MB buffer size friendly models (<a class=\"issue-link js-issue-link\" data-error-text=\"Failed to load title\" data-id=\"1989162122\" data-permission-text=\"Title is private\" data-url=\"https://github.com/mlc-ai/web-llm/issues/213\" data-hovercard-type=\"pull_request\" data-hovercard-url=\"/mlc-ai/web-llm/pull/213/hovercard\" href=\"https://github.com/mlc-ai/web-llm/pull/213\">mlc-ai#213</a>)"}}],"hasNextPage":false,"hasPreviousPage":false,"activityType":"all","actor":null,"timePeriod":"all","sort":"DESC","perPage":30,"cursor":"djE6ks8AAAADzuJMZgA","startCursor":null,"endCursor":null}},"title":"Activity · Navezjt/web-llm"}