TiGa-RCE/gte-Qwen2-1.5B-instruct-MLX-Q8
018
1{2 "schema_version": 1,3 "published_repository": "TiGa-RCE/gte-Qwen2-1.5B-instruct-MLX-Q8",4 "family": "gte-Qwen2-1.5B-instruct",5 "variant": "Q8",6 "upstream_repository": "Alibaba-NLP/gte-Qwen2-1.5B-instruct",7 "upstream_revision": "a9af15a6372d7d6b25e9fb07c2ccb9e1fe645644",8 "upstream_revision_evidence": "upstream revision verified at publication time; historical local snapshot metadata was not retained",9 "direct_parent": "TiGa-RCE/gte-Qwen2-1.5B-instruct-MLX-BF16",10 "direct_parent_weight_hashes": [11 {12 "file": "model.safetensors",13 "sha256": "9e9da58bd1371c47a08bc82f58bd29d33a6831094372dd717d68148f061dd11a",14 "bytes": 355243273215 }16 ],17 "conversion": {18 "method": "Q uniform affine quantization",19 "nominal_bits": 8,20 "group_size": 64,21 "importance_matrix": false,22 "importance_matrix_samples": null,23 "importance_matrix_sequence_length": null,24 "stack": {25 "omlx": "0.5.3",26 "mlx_lm": "0.31.3",27 "mlx": "0.32.0"28 },29 "lossy_parent": false30 },31 "weight_files": [32 {33 "file": "model.safetensors",34 "sha256": "58f2c49af27266301f16ee0724a976aa62c7b1a8ee654d44a509de97c68dff72",35 "bytes": 188742777636 }37 ],38 "evaluation": {39 "pair_count": 24,40 "top1": 1.0,41 "recall_at_5": 1.0,42 "mrr": 1.0,43 "mean_aligned_embedding_cosine_vs_bf16": 0.9996282458305359,44 "minimum_aligned_embedding_cosine_vs_bf16": 0.9993134140968323,45 "score_rmse_vs_bf16": 0.0022689001634716988,46 "queries_with_rank_change": 0,47 "gate_passed": true,48 "gate_criteria": {49 "top1_delta_min": 0.0,50 "recall_at_5_delta_min": 0.0,51 "mrr_delta_min": -0.01,52 "minimum_aligned_embedding_cosine_min": 0.99,53 "queries_with_rank_change_max": 254 }55 },56 "collection": "https://huggingface.co/collections/TiGa-RCE/mlx-embedding-quantization-matrix-q-oq-oqe-at-4-6-8-bit-6a68d11afb238d4fe967d70b"57}58 