Oct 14 19 days ago 17:44 docs: update model card, context lengths, and config specs Bartosz Hanc
Oct 12 21 days ago 12:35 update models Sep 7 15:10 Apply the model card standard 15:03 Add MLX int4 variants for 1B and 3B 13:54 Apply the model card standard 12:55 Apply the model card standard 12:46 Remove the QLoRA variants, unsupported since v0.10 The legacy compatibility layer pins them at the v0.8.0 tag, which is untouched, so they stay reachable for older library versions. Sep 3 14:48 Stop restating quantized, and drop the unused default quantized duplicated precisions.json, which is authoritative for the
quantized/non-quantized partition. default expressed a per-repository
recommendation nothing consumed: the library resolves its own per-platform
default. Both are gone from the schema. 12:17 Correct the published metadata Adds library_name to the model card, and pipeline_tag where it was missing.
Brings the config and card licenses into agreement, and repairs the variant
declarations that did not match the files actually published here. Aug 19 12:04 Update chat template for strict monotonicity and tool support Jun 22 09:43 Use min/max dynamic shape bounds for forward inputs/outputs 08:47 Add config.json for QLoRA variants 08:39 Backfill constant metadata method values in config.json from .pte May 14 08:34 Add stub root config.json for HF download counter May 13 19:37 Add spec-compliant config.json files 13:54 Add spec-compliant config.json files 13:24 Remove old-layout metadata orphaned by MODEL_SPEC.md restructure 13:11 Restructure to MODEL_SPEC.md convention Jan 28 11:29 Fix invalid JSON in config.json Add missing quotes around model_type key to make the JSON valid and parseable. May 8, 2025 12:54 Update README.md 12:53 Update README.md Apr 23, 2025 08:07 update non-quantized llama models to include context len in metadata Apr 17, 2025 07:56 update README 07:54 update README 07:45 update models to work with v0.6.0 runtime Apr 16, 2025 12:14 Update tokenizer_config pweglik
Apr 15, 2025 12:58 Upload tokenizer_config.json Apr 1, 2025 14:07 Replace .bin tokenizers with .json Nov 19, 2024 12:38 Add config.json to repository root 09:47 Add config.json to each of the models Nov 7, 2024 15:12 Update README.md 11:38 Upload llama3_2_bf16.pte 11:34 Upload llama3_2_3B_bf16.pte Nov 6, 2024 14:02 Delete llama3_2 13:37 Update README.md Nov 5, 2024 09:40 Delete llama-3.2-3B/spinquant/README.md 09:40 Delete llama-3.2-3B/original/README.md 09:40 Delete llama-3.2-3B/QLoRA/README.md 09:40 Delete llama-3.2-1B/README.md 09:40 Delete llama-3.2-1B/spinquant/README.md 09:40 Delete llama-3.2-1B/QLoRA/README.md 09:40 Delete llama-3.2-1B/original/README.md 09:39 Delete llama-3.2-3B/README.md Nov 4, 2024 15:02 Rename llama-3.2-3B/original/llama3_2_bf16.pte to llama-3.2-3B/original/llama3_2_3B_bf16.pte 15:02 Upload 2 files 14:58 Upload 2 files 14:52 Upload 2 files 14:49 Create original/README.md 14:49 Create spinquant/README.md 14:49 Create QLoRA/README.md 
