{ "format": "LLM-SCAN", "format_version": "1.0", "generator": { "name": "llmscan-engine", "version": "2.0.0" }, "scan_context": { "started_at": "2026-09-15T05:25:04+00:00Z", "ended_at": "2026-09-15T05:25:04+00:00Z", "duration_ms": 82480, "limits": { "timeout_ms": 60000, "sandbox": true }, "mode": "static-deep" }, "source": { "input_type": "huggingface", "filename": "gemma-4-E4B-it-SDFT_Heretic_RP.i1-IQ3_M.gguf", "size_bytes": 4714691264, "hashes": { "sha256": "434ed4a3938649780c23486849e2d62138d681e3628aa9396a1c83b0d5821f49" }, "repo_id": "mradermacher/gemma-4-E4B-it-SDFT_Heretic_RP-i1-GGUF", "file_sha": null, "repo_sha": "a6356be002df20b7c3abaa96b90b0ff328e2b0cf", "hf_provenance": { "repo_id": "mradermacher/gemma-4-E4B-it-SDFT_Heretic_RP-i1-GGUF", "repo_sha": "a6356be002df20b7c3abaa96b90b0ff328e2b0cf", "file_sha": null, "repo_license": "apache-2.0", "tags": [ "transformers", "gguf", "gemma4", "roleplay", "rp", "merged", "sdft", "gguf-ready", "en", "base_model:Ilya626/gemma-4-E4B-it-SDFT_Heretic_RP", "base_model:quantized:Ilya626/gemma-4-E4B-it-SDFT_Heretic_RP", "license:apache-2.0", "endpoints_compatible", "region:us", "imatrix", "conversational" ], "pipeline_tag": null, "downloads": 1156, "likes": 4, "has_readme": true, "has_config": false, "has_tokenizer_config": false, "siblings_count": 18, "siblings": [ { "name": ".gitattributes", "size": 2819, "sha": null }, { "name": "README.md", "size": 5362, "sha": null }, { "name": "gemma-4-E4B-it-SDFT_Heretic_RP.i1-IQ3_M.gguf", "size": 4714691264, "sha": null }, { "name": "gemma-4-E4B-it-SDFT_Heretic_RP.i1-IQ3_S.gguf", "size": 4663163584, "sha": null }, { "name": "gemma-4-E4B-it-SDFT_Heretic_RP.i1-IQ4_NL.gguf", "size": 5193953984, "sha": null }, { "name": "gemma-4-E4B-it-SDFT_Heretic_RP.i1-IQ4_XS.gguf", "size": 5070951104, "sha": null }, { "name": "gemma-4-E4B-it-SDFT_Heretic_RP.i1-Q2_K.gguf", "size": 4401316544, "sha": null }, { "name": "gemma-4-E4B-it-SDFT_Heretic_RP.i1-Q3_K_L.gguf", "size": 5021276864, "sha": null }, { "name": "gemma-4-E4B-it-SDFT_Heretic_RP.i1-Q3_K_M.gguf", "size": 4850391744, "sha": null }, { "name": "gemma-4-E4B-it-SDFT_Heretic_RP.i1-Q3_K_S.gguf", "size": 4654633664, "sha": null }, { "name": "gemma-4-E4B-it-SDFT_Heretic_RP.i1-Q4_0.gguf", "size": 5194117824, "sha": null }, { "name": "gemma-4-E4B-it-SDFT_Heretic_RP.i1-Q4_1.gguf", "size": 5435945664, "sha": null }, { "name": "gemma-4-E4B-it-SDFT_Heretic_RP.i1-Q4_K_M.gguf", "size": 5335286464, "sha": null }, { "name": "gemma-4-E4B-it-SDFT_Heretic_RP.i1-Q4_K_S.gguf", "size": 5202965184, "sha": null }, { "name": "gemma-4-E4B-it-SDFT_Heretic_RP.i1-Q5_K_M.gguf", "size": 5762908864, "sha": null }, { "name": "gemma-4-E4B-it-SDFT_Heretic_RP.i1-Q5_K_S.gguf", "size": 5685965504, "sha": null }, { "name": "gemma-4-E4B-it-SDFT_Heretic_RP.i1-Q6_K.gguf", "size": 6217257664, "sha": null }, { "name": "gemma-4-E4B-it-SDFT_Heretic_RP.imatrix.gguf", "size": 4429024, "sha": null } ], "trust_score": 50 } }, "model": { "format": "gguf", "arch": "gemma4", "quantization": "Q6_K", "params_estimate": 7518069290, "file_size_bytes": 4714691264, "sha256": "434ed4a3938649780c23486849e2d62138d681e3628aa9396a1c83b0d5821f49", "sha512": "094b9356abfe1d657d62f3df1bfc2ce340b6534e5b7739bf4fc5e81004081ede25c1ab15fd173f0a58086de6d6b88623c204dc493e98ec811bc69346584d243b", "kv_meta": { "values_redacted": true, "kv_count": 59, "keys": [ "gemma4.attention.head_count", "gemma4.attention.head_count_kv", "gemma4.attention.key_length", "gemma4.attention.key_length_swa", "gemma4.attention.layer_norm_rms_epsilon", "gemma4.attention.shared_kv_layers", "gemma4.attention.sliding_window", "gemma4.attention.sliding_window_pattern", "gemma4.attention.value_length", "gemma4.attention.value_length_swa", "gemma4.block_count", "gemma4.context_length", "gemma4.embedding_length", "gemma4.embedding_length_per_layer_input", "gemma4.feed_forward_length", "gemma4.final_logit_softcapping", "gemma4.rope.dimension_count", "gemma4.rope.dimension_count_swa", "gemma4.rope.freq_base", "gemma4.rope.freq_base_swa", "general.architecture", "general.base_model.0.name", "general.base_model.0.organization", "general.base_model.0.repo_url", "general.base_model.count", "general.file_type", "general.license", "general.name", "general.quantization_version", "general.sampling.temp", "general.sampling.top_k", "general.sampling.top_p", "general.size_label", "general.source.url", "general.tags", "general.type", "general.url", "mradermacher.convert_type", "mradermacher.quantize_version", "mradermacher.quantized_at", "mradermacher.quantized_by", "mradermacher.quantized_on", "quantize.imatrix.chunks_count", "quantize.imatrix.dataset", "quantize.imatrix.entries_count", "quantize.imatrix.file", "tokenizer.chat_template", "tokenizer.ggml.add_bos_token", "tokenizer.ggml.add_space_prefix", "tokenizer.ggml.bos_token_id", "tokenizer.ggml.eos_token_id", "tokenizer.ggml.mask_token_id", "tokenizer.ggml.merges", "tokenizer.ggml.model", "tokenizer.ggml.padding_token_id", "tokenizer.ggml.scores", "tokenizer.ggml.token_type", "tokenizer.ggml.tokens", "tokenizer.ggml.unknown_token_id" ] }, "scanner_model_info": { "magic": "GGUF", "version": 3, "architecture": "gemma4", "model_name": "Gemma 4 E4B It SDFT_Heretic_RP", "finetune": null, "basename": null, "size_label": "7.5B", "quantization": "Q6_K", "license": "apache-2.0", "context_length": 131072, "embedding_length": 2560, "feed_forward_length": 10240, "block_count": 42, "head_count": 8, "head_count_kv": 2, "vocab_size": 262144, "bos_token_id": 2, "eos_token_id": 1, "tokenizer_model": "gemma4", "tensor_count": 720, "param_estimate": 7518069290, "file_size_bytes": 4714691264, "file_size_mb": 4496.28, "sha256": "434ed4a3938649780c23486849e2d62138d681e3628aa9396a1c83b0d5821f49", "sha512": "094b9356abfe1d657d62f3df1bfc2ce340b6534e5b7739bf4fc5e81004081ede25c1ab15fd173f0a58086de6d6b88623c204dc493e98ec811bc69346584d243b", "alignment": 32, "data_section_start": 15820160, "header_end_offset": 15820160, "has_chat_template": true, "overlapping_tensor_pairs": 0, "misaligned_tensors": 0, "oob_tensor_count": 0, "dtype_histogram": { "F32": 339, "F16": 1, "Q6_K": 2, "IQ3_S": 289, "Q4_K": 89 }, "dtype_percentage": { "F32": 47.1, "F16": 0.1, "Q6_K": 0.3, "IQ3_S": 40.1, "Q4_K": 12.4 }, "top_tensors_by_size": [ { "name": "per_layer_token_embd.weight", "dtype": "Q6_K", "dims": [ 10752, 262144 ], "byte_len": 2312110080, "n_elem": 2818572288 }, { "name": "token_embd.weight", "dtype": "Q6_K", "dims": [ 2560, 262144 ], "byte_len": 550502400, "n_elem": 671088640 }, { "name": "per_layer_model_proj.weight", "dtype": "F16", "dims": [ 2560, 10752 ], "byte_len": 55050240, "n_elem": 27525120 }, { "name": "blk.0.ffn_down.weight", "dtype": "Q4_K", "dims": [ 10240, 2560 ], "byte_len": 14745600, "n_elem": 26214400 }, { "name": "blk.1.ffn_down.weight", "dtype": "Q4_K", "dims": [ 10240, 2560 ], "byte_len": 14745600, "n_elem": 26214400 }, { "name": "blk.2.ffn_down.weight", "dtype": "Q4_K", "dims": [ 10240, 2560 ], "byte_len": 14745600, "n_elem": 26214400 }, { "name": "blk.3.ffn_down.weight", "dtype": "Q4_K", "dims": [ 10240, 2560 ], "byte_len": 14745600, "n_elem": 26214400 }, { "name": "blk.4.ffn_down.weight", "dtype": "Q4_K", "dims": [ 10240, 2560 ], "byte_len": 14745600, "n_elem": 26214400 }, { "name": "blk.0.ffn_gate.weight", "dtype": "IQ3_S", "dims": [ 2560, 10240 ], "byte_len": 11264000, "n_elem": 26214400 }, { "name": "blk.0.ffn_up.weight", "dtype": "IQ3_S", "dims": [ 2560, 10240 ], "byte_len": 11264000, "n_elem": 26214400 } ], "tensor_bytes_sum": 4698871104, "tensor_delta_bytes": 15820160, "tensor_delta_mb": 15.09, "header_kv_dir_bytes": 15820160, "trailing_bytes": 0, "unaccounted_bytes": 0, "dtype_stats": { "F32": { "count": 339, "bytes": 2264384, "mb": 2.16, "avg_bits_per_elem": 32.016628 }, "F16": { "count": 1, "bytes": 55050240, "mb": 52.5, "avg_bits_per_elem": 16.0 }, "Q6_K": { "count": 2, "bytes": 2862612480, "mb": 2730.0, "avg_bits_per_elem": 6.5625 }, "IQ3_S": { "count": 289, "bytes": 1524582400, "mb": 1453.96, "avg_bits_per_elem": 3.4375 }, "Q4_K": { "count": 89, "bytes": 254361600, "mb": 242.58, "avg_bits_per_elem": 4.5 } }, "offsets_monotone": true, "max_tensor_offset": 4698589504, "tensor_preview": [ { "index": 0, "name": "output_norm.weight", "dims": [ 2560 ], "dtype_name": "F32", "dtype_id": 0, "offset": 0, "byte_len": 10240, "bits_per_elem": 32.0 }, { "index": 1, "name": "per_layer_model_proj.weight", "dims": [ 2560, 10752 ], "dtype_name": "F16", "dtype_id": 1, "offset": 10240, "byte_len": 55050240, "bits_per_elem": 16.0 }, { "index": 2, "name": "per_layer_proj_norm.weight", "dims": [ 256 ], "dtype_name": "F32", "dtype_id": 0, "offset": 55060480, "byte_len": 1024, "bits_per_elem": 32.0 }, { "index": 3, "name": "per_layer_token_embd.weight", "dims": [ 10752, 262144 ], "dtype_name": "Q6_K", "dtype_id": 14, "offset": 55061504, "byte_len": 2312110080, "bits_per_elem": 6.5625 }, { "index": 4, "name": "rope_freqs.weight", "dims": [ 256 ], "dtype_name": "F32", "dtype_id": 0, "offset": 2367171584, "byte_len": 1024, "bits_per_elem": 32.0 }, { "index": 5, "name": "token_embd.weight", "dims": [ 2560, 262144 ], "dtype_name": "Q6_K", "dtype_id": 14, "offset": 2367172608, "byte_len": 550502400, "bits_per_elem": 6.5625 }, { "index": 6, "name": "blk.0.attn_k.weight", "dims": [ 2560, 512 ], "dtype_name": "IQ3_S", "dtype_id": 21, "offset": 2917675008, "byte_len": 563200, "bits_per_elem": 3.4375 }, { "index": 7, "name": "blk.0.attn_k_norm.weight", "dims": [ 256 ], "dtype_name": "F32", "dtype_id": 0, "offset": 2918238208, "byte_len": 1024, "bits_per_elem": 32.0 }, { "index": 8, "name": "blk.0.attn_norm.weight", "dims": [ 2560 ], "dtype_name": "F32", "dtype_id": 0, "offset": 2918239232, "byte_len": 10240, "bits_per_elem": 32.0 }, { "index": 9, "name": "blk.0.attn_output.weight", "dims": [ 2048, 2560 ], "dtype_name": "Q4_K", "dtype_id": 12, "offset": 2918249472, "byte_len": 2949120, "bits_per_elem": 4.5 }, { "index": 10, "name": "blk.0.attn_q.weight", "dims": [ 2560, 2048 ], "dtype_name": "IQ3_S", "dtype_id": 21, "offset": 2921198592, "byte_len": 2252800, "bits_per_elem": 3.4375 }, { "index": 11, "name": "blk.0.attn_q_norm.weight", "dims": [ 256 ], "dtype_name": "F32", "dtype_id": 0, "offset": 2923451392, "byte_len": 1024, "bits_per_elem": 32.0 }, { "index": 12, "name": "blk.0.attn_v.weight", "dims": [ 2560, 512 ], "dtype_name": "Q4_K", "dtype_id": 12, "offset": 2923452416, "byte_len": 737280, "bits_per_elem": 4.5 }, { "index": 13, "name": "blk.0.ffn_down.weight", "dims": [ 10240, 2560 ], "dtype_name": "Q4_K", "dtype_id": 12, "offset": 2924189696, "byte_len": 14745600, "bits_per_elem": 4.5 }, { "index": 14, "name": "blk.0.ffn_gate.weight", "dims": [ 2560, 10240 ], "dtype_name": "IQ3_S", "dtype_id": 21, "offset": 2938935296, "byte_len": 11264000, "bits_per_elem": 3.4375 }, { "index": 15, "name": "blk.0.ffn_norm.weight", "dims": [ 2560 ], "dtype_name": "F32", "dtype_id": 0, "offset": 2950199296, "byte_len": 10240, "bits_per_elem": 32.0 }, { "index": 16, "name": "blk.0.ffn_up.weight", "dims": [ 2560, 10240 ], "dtype_name": "IQ3_S", "dtype_id": 21, "offset": 2950209536, "byte_len": 11264000, "bits_per_elem": 3.4375 }, { "index": 17, "name": "blk.0.inp_gate.weight", "dims": [ 2560, 256 ], "dtype_name": "IQ3_S", "dtype_id": 21, "offset": 2961473536, "byte_len": 281600, "bits_per_elem": 3.4375 }, { "index": 18, "name": "blk.0.layer_output_scale.weight", "dims": [ 1 ], "dtype_name": "F32", "dtype_id": 0, "offset": 2961755136, "byte_len": 32, "bits_per_elem": 256.0 }, { "index": 19, "name": "blk.0.post_attention_norm.weight", "dims": [ 2560 ], "dtype_name": "F32", "dtype_id": 0, "offset": 2961755168, "byte_len": 10240, "bits_per_elem": 32.0 }, { "index": 20, "name": "blk.0.post_ffw_norm.weight", "dims": [ 2560 ], "dtype_name": "F32", "dtype_id": 0, "offset": 2961765408, "byte_len": 10240, "bits_per_elem": 32.0 }, { "index": 21, "name": "blk.0.post_norm.weight", "dims": [ 2560 ], "dtype_name": "F32", "dtype_id": 0, "offset": 2961775648, "byte_len": 10240, "bits_per_elem": 32.0 }, { "index": 22, "name": "blk.0.proj.weight", "dims": [ 256, 2560 ], "dtype_name": "IQ3_S", "dtype_id": 21, "offset": 2961785888, "byte_len": 281600, "bits_per_elem": 3.4375 }, { "index": 23, "name": "blk.1.attn_k.weight", "dims": [ 2560, 512 ], "dtype_name": "IQ3_S", "dtype_id": 21, "offset": 2962067488, "byte_len": 563200, "bits_per_elem": 3.4375 }, { "index": 24, "name": "blk.1.attn_k_norm.weight", "dims": [ 256 ], "dtype_name": "F32", "dtype_id": 0, "offset": 2962630688, "byte_len": 1024, "bits_per_elem": 32.0 }, { "index": 25, "name": "blk.1.attn_norm.weight", "dims": [ 2560 ], "dtype_name": "F32", "dtype_id": 0, "offset": 2962631712, "byte_len": 10240, "bits_per_elem": 32.0 }, { "index": 26, "name": "blk.1.attn_output.weight", "dims": [ 2048, 2560 ], "dtype_name": "Q4_K", "dtype_id": 12, "offset": 2962641952, "byte_len": 2949120, "bits_per_elem": 4.5 }, { "index": 27, "name": "blk.1.attn_q.weight", "dims": [ 2560, 2048 ], "dtype_name": "IQ3_S", "dtype_id": 21, "offset": 2965591072, "byte_len": 2252800, "bits_per_elem": 3.4375 }, { "index": 28, "name": "blk.1.attn_q_norm.weight", "dims": [ 256 ], "dtype_name": "F32", "dtype_id": 0, "offset": 2967843872, "byte_len": 1024, "bits_per_elem": 32.0 }, { "index": 29, "name": "blk.1.attn_v.weight", "dims": [ 2560, 512 ], "dtype_name": "Q4_K", "dtype_id": 12, "offset": 2967844896, "byte_len": 737280, "bits_per_elem": 4.5 }, { "index": 30, "name": "blk.1.ffn_down.weight", "dims": [ 10240, 2560 ], "dtype_name": "Q4_K", "dtype_id": 12, "offset": 2968582176, "byte_len": 14745600, "bits_per_elem": 4.5 }, { "index": 31, "name": "blk.1.ffn_gate.weight", "dims": [ 2560, 10240 ], "dtype_name": "IQ3_S", "dtype_id": 21, "offset": 2983327776, "byte_len": 11264000, "bits_per_elem": 3.4375 }, { "index": 32, "name": "blk.1.ffn_norm.weight", "dims": [ 2560 ], "dtype_name": "F32", "dtype_id": 0, "offset": 2994591776, "byte_len": 10240, "bits_per_elem": 32.0 }, { "index": 33, "name": "blk.1.ffn_up.weight", "dims": [ 2560, 10240 ], "dtype_name": "IQ3_S", "dtype_id": 21, "offset": 2994602016, "byte_len": 11264000, "bits_per_elem": 3.4375 }, { "index": 34, "name": "blk.1.inp_gate.weight", "dims": [ 2560, 256 ], "dtype_name": "IQ3_S", "dtype_id": 21, "offset": 3005866016, "byte_len": 281600, "bits_per_elem": 3.4375 }, { "index": 35, "name": "blk.1.layer_output_scale.weight", "dims": [ 1 ], "dtype_name": "F32", "dtype_id": 0, "offset": 3006147616, "byte_len": 32, "bits_per_elem": 256.0 }, { "index": 36, "name": "blk.1.post_attention_norm.weight", "dims": [ 2560 ], "dtype_name": "F32", "dtype_id": 0, "offset": 3006147648, "byte_len": 10240, "bits_per_elem": 32.0 }, { "index": 37, "name": "blk.1.post_ffw_norm.weight", "dims": [ 2560 ], "dtype_name": "F32", "dtype_id": 0, "offset": 3006157888, "byte_len": 10240, "bits_per_elem": 32.0 }, { "index": 38, "name": "blk.1.post_norm.weight", "dims": [ 2560 ], "dtype_name": "F32", "dtype_id": 0, "offset": 3006168128, "byte_len": 10240, "bits_per_elem": 32.0 }, { "index": 39, "name": "blk.1.proj.weight", "dims": [ 256, 2560 ], "dtype_name": "IQ3_S", "dtype_id": 21, "offset": 3006178368, "byte_len": 281600, "bits_per_elem": 3.4375 }, { "index": 40, "name": "blk.2.attn_k.weight", "dims": [ 2560, 512 ], "dtype_name": "IQ3_S", "dtype_id": 21, "offset": 3006459968, "byte_len": 563200, "bits_per_elem": 3.4375 }, { "index": 41, "name": "blk.2.attn_k_norm.weight", "dims": [ 256 ], "dtype_name": "F32", "dtype_id": 0, "offset": 3007023168, "byte_len": 1024, "bits_per_elem": 32.0 }, { "index": 42, "name": "blk.2.attn_norm.weight", "dims": [ 2560 ], "dtype_name": "F32", "dtype_id": 0, "offset": 3007024192, "byte_len": 10240, "bits_per_elem": 32.0 }, { "index": 43, "name": "blk.2.attn_output.weight", "dims": [ 2048, 2560 ], "dtype_name": "Q4_K", "dtype_id": 12, "offset": 3007034432, "byte_len": 2949120, "bits_per_elem": 4.5 }, { "index": 44, "name": "blk.2.attn_q.weight", "dims": [ 2560, 2048 ], "dtype_name": "IQ3_S", "dtype_id": 21, "offset": 3009983552, "byte_len": 2252800, "bits_per_elem": 3.4375 }, { "index": 45, "name": "blk.2.attn_q_norm.weight", "dims": [ 256 ], "dtype_name": "F32", "dtype_id": 0, "offset": 3012236352, "byte_len": 1024, "bits_per_elem": 32.0 }, { "index": 46, "name": "blk.2.attn_v.weight", "dims": [ 2560, 512 ], "dtype_name": "Q4_K", "dtype_id": 12, "offset": 3012237376, "byte_len": 737280, "bits_per_elem": 4.5 }, { "index": 47, "name": "blk.2.ffn_down.weight", "dims": [ 10240, 2560 ], "dtype_name": "Q4_K", "dtype_id": 12, "offset": 3012974656, "byte_len": 14745600, "bits_per_elem": 4.5 }, { "index": 48, "name": "blk.2.ffn_gate.weight", "dims": [ 2560, 10240 ], "dtype_name": "IQ3_S", "dtype_id": 21, "offset": 3027720256, "byte_len": 11264000, "bits_per_elem": 3.4375 }, { "index": 49, "name": "blk.2.ffn_norm.weight", "dims": [ 2560 ], "dtype_name": "F32", "dtype_id": 0, "offset": 3038984256, "byte_len": 10240, "bits_per_elem": 32.0 } ], "chat_template": "{%- macro format_parameters(properties, required) -%}\n {%- set standard_keys = ['description', 'type', 'properties', 'required', 'nullable'] -%}\n {%- set ns = namespace(found_first=false) -%}\n {%- for key, value in properties | dictsort -%}\n {%- set add_comma = false -%}\n {%- if key not in standard_keys -%}\n {%- if ns.found_first %},{% endif -%}\n {%- set ns.found_first = true -%}\n {{ key }}:{\n {%- if value['description'] -%}\n description:<|\"|>{{ value['description'] }}<|\"|>\n {%- set add_comma = true -%}\n {%- endif -%}\n {%- if value['nullable'] %}\n {%- if add_comma %},{%- else -%} {%- set add_comma = true -%} {% endif -%}\n nullable:true\n {%- endif -%}\n {%- if value['type'] | upper == 'STRING' -%}\n {%- if value['enum'] -%}\n {%- if add_comma %},{%- else -%} {%- set add_comma = true -%} {% endif -%}\n enum:{{ format_argument(value['enum']) }}\n {%- endif -%}\n {%- elif value['type'] | upper == 'OBJECT' -%}\n ,properties:{\n {%- if value['properties'] is defined and value['properties'] is mapping -%}\n {{- format_parameters(value['properties'], value['required'] | default([])) -}}\n {%- elif value is mapping -%}\n {{- format_parameters(value, value['required'] | default([])) -}}\n {%- endif -%}\n }\n {%- if value['required'] -%}\n ,required:[\n {%- for item in value['required'] | default([]) -%}\n <|\"|>{{- item -}}<|\"|>\n {%- if not loop.last %},{% endif -%}\n {%- endfor -%}\n ]\n {%- endif -%}\n {%- elif value['type'] | upper == 'ARRAY' -%}\n {%- if value['items'] is mapping and value['items'] -%}\n ,items:{\n {%- set ns_items = namespace(found_first=false) -%}\n {%- for item_key, item_value in value['items'] | dictsort -%}\n {%- if item_value is not none -%}\n {%- if ns_items.found_first %},{% endif -%}\n {%- set ns_items.found_first = true -%}\n {%- if item_key == 'properties' -%}\n properties:{\n {%- if item_value is mapping -%}\n {{- format_parameters(item_value, value['items']['required'] | default([])) -}}\n {%- endif -%}\n }\n {%- elif item_key == 'required' -%}\n required:[\n {%- for req_item in item_value -%}\n <|\"|>{{- req_item -}}<|\"|>\n {%- if not loop.last %},{% endif -%}\n {%- endfor -%}\n ]\n {%- elif item_key == 'type' -%}\n {%- if item_value is string -%}\n type:{{ format_argument(item_value | upper) }}\n {%- else -%}\n type:{{ format_argument(item_value | map('upper') | list) }}\n {%- endif -%}\n {%- else -%}\n {{ item_key }}:{{ format_argument(item_value) }}\n {%- endif -%}\n {%- endif -%}\n {%- endfor -%}\n }\n {%- endif -%}\n {%- endif -%}\n {%- if add_comma %},{%- else -%} {%- set add_comma = true -%} {% endif -%}\n type:<|\"|>{{ value['type'] | upper }}<|\"|>}\n {%- endif -%}\n {%- endfor -%}\n{%- endmacro -%}\n{%- macro format_function_declaration(tool_data) -%}\n declaration:{{- tool_data['function']['name'] -}}{description:<|\"|>{{- tool_data['function']['description'] -}}<|\"|>\n {%- set params = tool_data['function']['parameters'] -%}\n {%- if params -%}\n ,parameters:{\n {%- if params['properties'] -%}\n properties:{ {{- format_parameters(params['properties'], params['required']) -}} },\n {%- endif -%}\n {%- if params['required'] -%}\n required:[\n {%- for item in params['required'] -%}\n <|\"|>{{- item -}}<|\"|>\n {{- ',' if not loop.last -}}\n {%- endfor -%}\n ],\n {%- endif -%}\n {%- if params['type'] -%}\n type:<|\"|>{{- params['type'] | upper -}}<|\"|>}\n {%- endif -%}\n {%- endif -%}\n {%- if 'response' in tool_data['function'] -%}\n {%- set response_declaration = tool_data['function']['response'] -%}\n ,response:{\n {%- if response_declaration['description'] -%}\n description:<|\"|>{{- response_declaration['description'] -}}<|\"|>,\n {%- endif -%}\n {%- if response_declaration['type'] | upper == 'OBJECT' -%}\n type:<|\"|>{{- response_declaration['type'] | upper -}}<|\"|>}\n {%- endif -%}\n {%- endif -%}\n }\n{%- endmacro -%}\n{%- macro format_argument(argument, escape_keys=True) -%}\n {%- if argument is string -%}\n {{- '<|\"|>' + argument + '<|\"|>' -}}\n {%- elif argument is boolean -%}\n {{- 'true' if argument else 'false' -}}\n {%- elif argument is mapping -%}\n {{- '{' -}}\n {%- set ns = namespace(found_first=false) -%}\n {%- for key, value in argument | dictsort -%}\n {%- if ns.found_first %},{% endif -%}\n {%- set ns.found_first = true -%}\n {%- if escape_keys -%}\n {{- '<|\"|>' + key + '<|\"|>' -}}\n {%- else -%}\n {{- key -}}\n {%- endif -%}\n :{{- format_argument(value, escape_keys=escape_keys) -}}\n {%- endfor -%}\n {{- '}' -}}\n {%- elif argument is sequence -%}\n {{- '[' -}}\n {%- for item in argument -%}\n {{- format_argument(item, escape_keys=escape_keys) -}}\n {%- if not loop.last %},{% endif -%}\n {%- endfor -%}\n {{- ']' -}}\n {%- else -%}\n {{- argument -}}\n {%- endif -%}\n{%- endmacro -%}\n{%- macro strip_thinking(text) -%}\n {%- set ns = namespace(result='') -%}\n {%- for part in text.split('') -%}\n {%- if '<|channel>' in part -%}\n {%- set ns.result = ns.result + part.split('<|channel>')[0] -%}\n {%- else -%}\n {%- set ns.result = ns.result + part -%}\n {%- endif -%}\n {%- endfor -%}\n {{- ns.result | trim -}}\n{%- endmacro -%}\n\n{%- set ns = namespace(prev_message_type=None) -%}\n{%- set loop_messages = messages -%}\n{{ bos_token }}\n{#- Handle System/Tool Definitions Block -#}\n{%- if (enable_thinking is defined and enable_thinking) or tools or messages[0]['role'] in ['system', 'developer'] -%}\n {{- '<|turn>system\\n' -}}\n\n {#- Inject Thinking token at the very top of the FIRST system turn -#}\n {%- if enable_thinking is defined and enable_thinking -%}\n {{- '<|think|>' -}}\n {%- set ns.prev_message_type = 'think' -%}\n {%- endif -%}\n\n {%- if messages[0]['role'] in ['system', 'developer'] -%}\n {{- messages[0]['content'] | trim -}}\n {%- set loop_messages = messages[1:] -%}\n {%- endif -%}\n\n {%- if tools -%}\n {%- for tool in tools %}\n {{- '<|tool>' -}}\n {{- format_function_declaration(tool) | trim -}}\n {{- '' -}}\n {%- endfor %}\n {%- set ns.prev_message_type = 'tool' -%}\n {%- endif -%}\n\n {{- '\\n' -}}\n{%- endif %}\n\n{#- Loop through messages -#}\n{%- for message in loop_messages -%}\n {%- set ns.prev_message_type = None -%}\n {%- set role = 'model' if message['role'] == 'assistant' else message['role'] -%}\n {{- '<|turn>' + role + '\\n' }}\n\n {%- if message['tool_calls'] -%}\n {%- for tool_call in message['tool_calls'] -%}\n {%- set function = tool_call['function'] -%}\n {{- '<|tool_call>call:' + function['name'] + '{' -}}\n {%- if function['arguments'] is mapping -%}\n {%- set ns_args = namespace(found_first=false) -%}\n {%- for key, value in function['arguments'] | dictsort -%}\n {%- if ns_args.found_first %},{% endif -%}\n {%- set ns_args.found_first = true -%}\n {{- key -}}:{{- format_argument(value, escape_keys=False) -}}\n {%- endfor -%}\n {%- elif function['arguments'] is string -%}\n {{- function['arguments'] -}}\n {%- endif -%}\n {{- '}' -}}\n {%- endfor -%}\n {%- set ns.prev_message_type = 'tool_call' -%}\n {%- endif -%}\n\n {%- if message['tool_responses'] -%}\n {#- Tool Response handling -#}\n {%- for tool_response in message['tool_responses'] -%}\n {{- '<|tool_response>' -}}\n {%- if tool_response['response'] is mapping -%}\n {{- 'response:' + tool_response['name'] | default('unknown') + '{' -}}\n {%- for key, value in tool_response['response'] | dictsort -%}\n {{- key -}}:{{- format_argument(value, escape_keys=False) -}}\n {%- if not loop.last %},{% endif -%}\n {%- endfor -%}\n {{- '}' -}}\n {%- else -%}\n {{- 'response:' + tool_response['name'] | default('unknown') + '{value:' + format_argument(tool_response['response'], escape_keys=False) + '}' -}}\n {%- endif -%}\n {{- '' -}}\n {%- endfor -%}\n {%- set ns.prev_message_type = 'tool_response' -%}\n {%- endif -%}\n\n {%- if message['content'] is string -%}\n {%- if role == 'model' -%}\n {{- strip_thinking(message['content']) -}}\n {%- else -%}\n {{- message['content'] | trim -}}\n {%- endif -%}\n {%- elif message['content'] is sequence -%}\n {%- for item in message['content'] -%}\n {%- if item['type'] == 'text' -%}\n {%- if role == 'model' -%}\n {{- strip_thinking(item['text']) -}}\n {%- else -%}\n {{- item['text'] | trim -}}\n {%- endif -%}\n {%- elif item['type'] == 'image' -%}\n {{- '\\n\\n<|image|>\\n\\n' -}}\n {%- set ns.prev_message_type = 'image' -%}\n {%- elif item['type'] == 'audio' -%}\n {{- '<|audio|>' -}}\n {%- set ns.prev_message_type = 'audio' -%}\n {%- elif item['type'] == 'video' -%}\n {{- '\\n\\n<|video|>\\n\\n' -}}\n {%- set ns.prev_message_type = 'video' -%}\n {%- endif -%}\n {%- endfor -%}\n {%- endif -%}\n\n {%- if not (message['tool_responses'] and not message['content']) -%}\n {{- '\\n' -}}\n {%- endif -%}\n{%- endfor -%}\n\n{%- if add_generation_prompt -%}\n {%- if ns.prev_message_type != 'tool_response' -%}\n {{- '<|turn>model\\n' -}}\n {%- endif -%}\n{%- endif -%}", "kv_count": 59, "kv_keys": [ "gemma4.attention.head_count", "gemma4.attention.head_count_kv", "gemma4.attention.key_length", "gemma4.attention.key_length_swa", "gemma4.attention.layer_norm_rms_epsilon", "gemma4.attention.shared_kv_layers", "gemma4.attention.sliding_window", "gemma4.attention.sliding_window_pattern", "gemma4.attention.value_length", "gemma4.attention.value_length_swa", "gemma4.block_count", "gemma4.context_length", "gemma4.embedding_length", "gemma4.embedding_length_per_layer_input", "gemma4.feed_forward_length", "gemma4.final_logit_softcapping", "gemma4.rope.dimension_count", "gemma4.rope.dimension_count_swa", "gemma4.rope.freq_base", "gemma4.rope.freq_base_swa", "general.architecture", "general.base_model.0.name", "general.base_model.0.organization", "general.base_model.0.repo_url", "general.base_model.count", "general.file_type", "general.license", "general.name", "general.quantization_version", "general.sampling.temp", "general.sampling.top_k", "general.sampling.top_p", "general.size_label", "general.source.url", "general.tags", "general.type", "general.url", "mradermacher.convert_type", "mradermacher.quantize_version", "mradermacher.quantized_at", "mradermacher.quantized_by", "mradermacher.quantized_on", "quantize.imatrix.chunks_count", "quantize.imatrix.dataset", "quantize.imatrix.entries_count", "quantize.imatrix.file", "tokenizer.chat_template", "tokenizer.ggml.add_bos_token", "tokenizer.ggml.add_space_prefix", "tokenizer.ggml.bos_token_id", "tokenizer.ggml.eos_token_id", "tokenizer.ggml.mask_token_id", "tokenizer.ggml.merges", "tokenizer.ggml.model", "tokenizer.ggml.padding_token_id", "tokenizer.ggml.scores", "tokenizer.ggml.token_type", "tokenizer.ggml.tokens", "tokenizer.ggml.unknown_token_id" ], "quantization_declared": "IQ3_S" }, "gguf_version": 3, "architecture": "gemma4", "model_name": "Gemma 4 E4B It SDFT_Heretic_RP", "finetune": null, "basename": null, "size_label": "7.5B", "license": "apache-2.0", "context_length": 131072, "embedding_length": 2560, "feed_forward_length": 10240, "block_count": 42, "head_count": 8, "head_count_kv": 2, "vocab_size": 262144, "bos_token_id": 2, "eos_token_id": 1, "tokenizer_model": "gemma4", "tensor_count": 720, "param_estimate": 7518069290, "file_size_mb": 4496.28, "has_chat_template": true, "alignment": 32, "overlapping_tensor_pairs": 0, "misaligned_tensors": 0, "oob_tensor_count": 0, "dtype_histogram": { "F32": 339, "F16": 1, "Q6_K": 2, "IQ3_S": 289, "Q4_K": 89 }, "dtype_percentage": { "F32": 47.1, "F16": 0.1, "Q6_K": 0.3, "IQ3_S": 40.1, "Q4_K": 12.4 }, "top_tensors_by_size": [ { "name": "per_layer_token_embd.weight", "dtype": "Q6_K", "dims": [ 10752, 262144 ], "byte_len": 2312110080, "n_elem": 2818572288 }, { "name": "token_embd.weight", "dtype": "Q6_K", "dims": [ 2560, 262144 ], "byte_len": 550502400, "n_elem": 671088640 }, { "name": "per_layer_model_proj.weight", "dtype": "F16", "dims": [ 2560, 10752 ], "byte_len": 55050240, "n_elem": 27525120 }, { "name": "blk.0.ffn_down.weight", "dtype": "Q4_K", "dims": [ 10240, 2560 ], "byte_len": 14745600, "n_elem": 26214400 }, { "name": "blk.1.ffn_down.weight", "dtype": "Q4_K", "dims": [ 10240, 2560 ], "byte_len": 14745600, "n_elem": 26214400 }, { "name": "blk.2.ffn_down.weight", "dtype": "Q4_K", "dims": [ 10240, 2560 ], "byte_len": 14745600, "n_elem": 26214400 }, { "name": "blk.3.ffn_down.weight", "dtype": "Q4_K", "dims": [ 10240, 2560 ], "byte_len": 14745600, "n_elem": 26214400 }, { "name": "blk.4.ffn_down.weight", "dtype": "Q4_K", "dims": [ 10240, 2560 ], "byte_len": 14745600, "n_elem": 26214400 }, { "name": "blk.0.ffn_gate.weight", "dtype": "IQ3_S", "dims": [ 2560, 10240 ], "byte_len": 11264000, "n_elem": 26214400 }, { "name": "blk.0.ffn_up.weight", "dtype": "IQ3_S", "dims": [ 2560, 10240 ], "byte_len": 11264000, "n_elem": 26214400 } ], "tensor_bytes_sum": 4698871104, "tensor_delta_bytes": 15820160, "tensor_delta_mb": 15.09, "header_kv_dir_bytes": 15820160, "trailing_bytes": 0, "unaccounted_bytes": 0, "dtype_stats": { "F32": { "count": 339, "bytes": 2264384, "mb": 2.16, "avg_bits_per_elem": 32.016628 }, "F16": { "count": 1, "bytes": 55050240, "mb": 52.5, "avg_bits_per_elem": 16.0 }, "Q6_K": { "count": 2, "bytes": 2862612480, "mb": 2730.0, "avg_bits_per_elem": 6.5625 }, "IQ3_S": { "count": 289, "bytes": 1524582400, "mb": 1453.96, "avg_bits_per_elem": 3.4375 }, "Q4_K": { "count": 89, "bytes": 254361600, "mb": 242.58, "avg_bits_per_elem": 4.5 } }, "offsets_monotone": true, "max_tensor_offset": 4698589504, "tensor_preview": [ { "index": 0, "name": "output_norm.weight", "dims": [ 2560 ], "dtype_name": "F32", "dtype_id": 0, "offset": 0, "byte_len": 10240, "bits_per_elem": 32.0 }, { "index": 1, "name": "per_layer_model_proj.weight", "dims": [ 2560, 10752 ], "dtype_name": "F16", "dtype_id": 1, "offset": 10240, "byte_len": 55050240, "bits_per_elem": 16.0 }, { "index": 2, "name": "per_layer_proj_norm.weight", "dims": [ 256 ], "dtype_name": "F32", "dtype_id": 0, "offset": 55060480, "byte_len": 1024, "bits_per_elem": 32.0 }, { "index": 3, "name": "per_layer_token_embd.weight", "dims": [ 10752, 262144 ], "dtype_name": "Q6_K", "dtype_id": 14, "offset": 55061504, "byte_len": 2312110080, "bits_per_elem": 6.5625 }, { "index": 4, "name": "rope_freqs.weight", "dims": [ 256 ], "dtype_name": "F32", "dtype_id": 0, "offset": 2367171584, "byte_len": 1024, "bits_per_elem": 32.0 }, { "index": 5, "name": "token_embd.weight", "dims": [ 2560, 262144 ], "dtype_name": "Q6_K", "dtype_id": 14, "offset": 2367172608, "byte_len": 550502400, "bits_per_elem": 6.5625 }, { "index": 6, "name": "blk.0.attn_k.weight", "dims": [ 2560, 512 ], "dtype_name": "IQ3_S", "dtype_id": 21, "offset": 2917675008, "byte_len": 563200, "bits_per_elem": 3.4375 }, { "index": 7, "name": "blk.0.attn_k_norm.weight", "dims": [ 256 ], "dtype_name": "F32", "dtype_id": 0, "offset": 2918238208, "byte_len": 1024, "bits_per_elem": 32.0 }, { "index": 8, "name": "blk.0.attn_norm.weight", "dims": [ 2560 ], "dtype_name": "F32", "dtype_id": 0, "offset": 2918239232, "byte_len": 10240, "bits_per_elem": 32.0 }, { "index": 9, "name": "blk.0.attn_output.weight", "dims": [ 2048, 2560 ], "dtype_name": "Q4_K", "dtype_id": 12, "offset": 2918249472, "byte_len": 2949120, "bits_per_elem": 4.5 }, { "index": 10, "name": "blk.0.attn_q.weight", "dims": [ 2560, 2048 ], "dtype_name": "IQ3_S", "dtype_id": 21, "offset": 2921198592, "byte_len": 2252800, "bits_per_elem": 3.4375 }, { "index": 11, "name": "blk.0.attn_q_norm.weight", "dims": [ 256 ], "dtype_name": "F32", "dtype_id": 0, "offset": 2923451392, "byte_len": 1024, "bits_per_elem": 32.0 }, { "index": 12, "name": "blk.0.attn_v.weight", "dims": [ 2560, 512 ], "dtype_name": "Q4_K", "dtype_id": 12, "offset": 2923452416, "byte_len": 737280, "bits_per_elem": 4.5 }, { "index": 13, "name": "blk.0.ffn_down.weight", "dims": [ 10240, 2560 ], "dtype_name": "Q4_K", "dtype_id": 12, "offset": 2924189696, "byte_len": 14745600, "bits_per_elem": 4.5 }, { "index": 14, "name": "blk.0.ffn_gate.weight", "dims": [ 2560, 10240 ], "dtype_name": "IQ3_S", "dtype_id": 21, "offset": 2938935296, "byte_len": 11264000, "bits_per_elem": 3.4375 }, { "index": 15, "name": "blk.0.ffn_norm.weight", "dims": [ 2560 ], "dtype_name": "F32", "dtype_id": 0, "offset": 2950199296, "byte_len": 10240, "bits_per_elem": 32.0 }, { "index": 16, "name": "blk.0.ffn_up.weight", "dims": [ 2560, 10240 ], "dtype_name": "IQ3_S", "dtype_id": 21, "offset": 2950209536, "byte_len": 11264000, "bits_per_elem": 3.4375 }, { "index": 17, "name": "blk.0.inp_gate.weight", "dims": [ 2560, 256 ], "dtype_name": "IQ3_S", "dtype_id": 21, "offset": 2961473536, "byte_len": 281600, "bits_per_elem": 3.4375 }, { "index": 18, "name": "blk.0.layer_output_scale.weight", "dims": [ 1 ], "dtype_name": "F32", "dtype_id": 0, "offset": 2961755136, "byte_len": 32, "bits_per_elem": 256.0 }, { "index": 19, "name": "blk.0.post_attention_norm.weight", "dims": [ 2560 ], "dtype_name": "F32", "dtype_id": 0, "offset": 2961755168, "byte_len": 10240, "bits_per_elem": 32.0 }, { "index": 20, "name": "blk.0.post_ffw_norm.weight", "dims": [ 2560 ], "dtype_name": "F32", "dtype_id": 0, "offset": 2961765408, "byte_len": 10240, "bits_per_elem": 32.0 }, { "index": 21, "name": "blk.0.post_norm.weight", "dims": [ 2560 ], "dtype_name": "F32", "dtype_id": 0, "offset": 2961775648, "byte_len": 10240, "bits_per_elem": 32.0 }, { "index": 22, "name": "blk.0.proj.weight", "dims": [ 256, 2560 ], "dtype_name": "IQ3_S", "dtype_id": 21, "offset": 2961785888, "byte_len": 281600, "bits_per_elem": 3.4375 }, { "index": 23, "name": "blk.1.attn_k.weight", "dims": [ 2560, 512 ], "dtype_name": "IQ3_S", "dtype_id": 21, "offset": 2962067488, "byte_len": 563200, "bits_per_elem": 3.4375 }, { "index": 24, "name": "blk.1.attn_k_norm.weight", "dims": [ 256 ], "dtype_name": "F32", "dtype_id": 0, "offset": 2962630688, "byte_len": 1024, "bits_per_elem": 32.0 }, { "index": 25, "name": "blk.1.attn_norm.weight", "dims": [ 2560 ], "dtype_name": "F32", "dtype_id": 0, "offset": 2962631712, "byte_len": 10240, "bits_per_elem": 32.0 }, { "index": 26, "name": "blk.1.attn_output.weight", "dims": [ 2048, 2560 ], "dtype_name": "Q4_K", "dtype_id": 12, "offset": 2962641952, "byte_len": 2949120, "bits_per_elem": 4.5 }, { "index": 27, "name": "blk.1.attn_q.weight", "dims": [ 2560, 2048 ], "dtype_name": "IQ3_S", "dtype_id": 21, "offset": 2965591072, "byte_len": 2252800, "bits_per_elem": 3.4375 }, { "index": 28, "name": "blk.1.attn_q_norm.weight", "dims": [ 256 ], "dtype_name": "F32", "dtype_id": 0, "offset": 2967843872, "byte_len": 1024, "bits_per_elem": 32.0 }, { "index": 29, "name": "blk.1.attn_v.weight", "dims": [ 2560, 512 ], "dtype_name": "Q4_K", "dtype_id": 12, "offset": 2967844896, "byte_len": 737280, "bits_per_elem": 4.5 }, { "index": 30, "name": "blk.1.ffn_down.weight", "dims": [ 10240, 2560 ], "dtype_name": "Q4_K", "dtype_id": 12, "offset": 2968582176, "byte_len": 14745600, "bits_per_elem": 4.5 }, { "index": 31, "name": "blk.1.ffn_gate.weight", "dims": [ 2560, 10240 ], "dtype_name": "IQ3_S", "dtype_id": 21, "offset": 2983327776, "byte_len": 11264000, "bits_per_elem": 3.4375 }, { "index": 32, "name": "blk.1.ffn_norm.weight", "dims": [ 2560 ], "dtype_name": "F32", "dtype_id": 0, "offset": 2994591776, "byte_len": 10240, "bits_per_elem": 32.0 }, { "index": 33, "name": "blk.1.ffn_up.weight", "dims": [ 2560, 10240 ], "dtype_name": "IQ3_S", "dtype_id": 21, "offset": 2994602016, "byte_len": 11264000, "bits_per_elem": 3.4375 }, { "index": 34, "name": "blk.1.inp_gate.weight", "dims": [ 2560, 256 ], "dtype_name": "IQ3_S", "dtype_id": 21, "offset": 3005866016, "byte_len": 281600, "bits_per_elem": 3.4375 }, { "index": 35, "name": "blk.1.layer_output_scale.weight", "dims": [ 1 ], "dtype_name": "F32", "dtype_id": 0, "offset": 3006147616, "byte_len": 32, "bits_per_elem": 256.0 }, { "index": 36, "name": "blk.1.post_attention_norm.weight", "dims": [ 2560 ], "dtype_name": "F32", "dtype_id": 0, "offset": 3006147648, "byte_len": 10240, "bits_per_elem": 32.0 }, { "index": 37, "name": "blk.1.post_ffw_norm.weight", "dims": [ 2560 ], "dtype_name": "F32", "dtype_id": 0, "offset": 3006157888, "byte_len": 10240, "bits_per_elem": 32.0 }, { "index": 38, "name": "blk.1.post_norm.weight", "dims": [ 2560 ], "dtype_name": "F32", "dtype_id": 0, "offset": 3006168128, "byte_len": 10240, "bits_per_elem": 32.0 }, { "index": 39, "name": "blk.1.proj.weight", "dims": [ 256, 2560 ], "dtype_name": "IQ3_S", "dtype_id": 21, "offset": 3006178368, "byte_len": 281600, "bits_per_elem": 3.4375 }, { "index": 40, "name": "blk.2.attn_k.weight", "dims": [ 2560, 512 ], "dtype_name": "IQ3_S", "dtype_id": 21, "offset": 3006459968, "byte_len": 563200, "bits_per_elem": 3.4375 }, { "index": 41, "name": "blk.2.attn_k_norm.weight", "dims": [ 256 ], "dtype_name": "F32", "dtype_id": 0, "offset": 3007023168, "byte_len": 1024, "bits_per_elem": 32.0 }, { "index": 42, "name": "blk.2.attn_norm.weight", "dims": [ 2560 ], "dtype_name": "F32", "dtype_id": 0, "offset": 3007024192, "byte_len": 10240, "bits_per_elem": 32.0 }, { "index": 43, "name": "blk.2.attn_output.weight", "dims": [ 2048, 2560 ], "dtype_name": "Q4_K", "dtype_id": 12, "offset": 3007034432, "byte_len": 2949120, "bits_per_elem": 4.5 }, { "index": 44, "name": "blk.2.attn_q.weight", "dims": [ 2560, 2048 ], "dtype_name": "IQ3_S", "dtype_id": 21, "offset": 3009983552, "byte_len": 2252800, "bits_per_elem": 3.4375 }, { "index": 45, "name": "blk.2.attn_q_norm.weight", "dims": [ 256 ], "dtype_name": "F32", "dtype_id": 0, "offset": 3012236352, "byte_len": 1024, "bits_per_elem": 32.0 }, { "index": 46, "name": "blk.2.attn_v.weight", "dims": [ 2560, 512 ], "dtype_name": "Q4_K", "dtype_id": 12, "offset": 3012237376, "byte_len": 737280, "bits_per_elem": 4.5 }, { "index": 47, "name": "blk.2.ffn_down.weight", "dims": [ 10240, 2560 ], "dtype_name": "Q4_K", "dtype_id": 12, "offset": 3012974656, "byte_len": 14745600, "bits_per_elem": 4.5 }, { "index": 48, "name": "blk.2.ffn_gate.weight", "dims": [ 2560, 10240 ], "dtype_name": "IQ3_S", "dtype_id": 21, "offset": 3027720256, "byte_len": 11264000, "bits_per_elem": 3.4375 }, { "index": 49, "name": "blk.2.ffn_norm.weight", "dims": [ 2560 ], "dtype_name": "F32", "dtype_id": 0, "offset": 3038984256, "byte_len": 10240, "bits_per_elem": 32.0 } ], "chat_template": "{%- macro format_parameters(properties, required) -%}\n {%- set standard_keys = ['description', 'type', 'properties', 'required', 'nullable'] -%}\n {%- set ns = namespace(found_first=false) -%}\n {%- for key, value in properties | dictsort -%}\n {%- set add_comma = false -%}\n {%- if key not in standard_keys -%}\n {%- if ns.found_first %},{% endif -%}\n {%- set ns.found_first = true -%}\n {{ key }}:{\n {%- if value['description'] -%}\n description:<|\"|>{{ value['description'] }}<|\"|>\n {%- set add_comma = true -%}\n {%- endif -%}\n {%- if value['nullable'] %}\n {%- if add_comma %},{%- else -%} {%- set add_comma = true -%} {% endif -%}\n nullable:true\n {%- endif -%}\n {%- if value['type'] | upper == 'STRING' -%}\n {%- if value['enum'] -%}\n {%- if add_comma %},{%- else -%} {%- set add_comma = true -%} {% endif -%}\n enum:{{ format_argument(value['enum']) }}\n {%- endif -%}\n {%- elif value['type'] | upper == 'OBJECT' -%}\n ,properties:{\n {%- if value['properties'] is defined and value['properties'] is mapping -%}\n {{- format_parameters(value['properties'], value['required'] | default([])) -}}\n {%- elif value is mapping -%}\n {{- format_parameters(value, value['required'] | default([])) -}}\n {%- endif -%}\n }\n {%- if value['required'] -%}\n ,required:[\n {%- for item in value['required'] | default([]) -%}\n <|\"|>{{- item -}}<|\"|>\n {%- if not loop.last %},{% endif -%}\n {%- endfor -%}\n ]\n {%- endif -%}\n {%- elif value['type'] | upper == 'ARRAY' -%}\n {%- if value['items'] is mapping and value['items'] -%}\n ,items:{\n {%- set ns_items = namespace(found_first=false) -%}\n {%- for item_key, item_value in value['items'] | dictsort -%}\n {%- if item_value is not none -%}\n {%- if ns_items.found_first %},{% endif -%}\n {%- set ns_items.found_first = true -%}\n {%- if item_key == 'properties' -%}\n properties:{\n {%- if item_value is mapping -%}\n {{- format_parameters(item_value, value['items']['required'] | default([])) -}}\n {%- endif -%}\n }\n {%- elif item_key == 'required' -%}\n required:[\n {%- for req_item in item_value -%}\n <|\"|>{{- req_item -}}<|\"|>\n {%- if not loop.last %},{% endif -%}\n {%- endfor -%}\n ]\n {%- elif item_key == 'type' -%}\n {%- if item_value is string -%}\n type:{{ format_argument(item_value | upper) }}\n {%- else -%}\n type:{{ format_argument(item_value | map('upper') | list) }}\n {%- endif -%}\n {%- else -%}\n {{ item_key }}:{{ format_argument(item_value) }}\n {%- endif -%}\n {%- endif -%}\n {%- endfor -%}\n }\n {%- endif -%}\n {%- endif -%}\n {%- if add_comma %},{%- else -%} {%- set add_comma = true -%} {% endif -%}\n type:<|\"|>{{ value['type'] | upper }}<|\"|>}\n {%- endif -%}\n {%- endfor -%}\n{%- endmacro -%}\n{%- macro format_function_declaration(tool_data) -%}\n declaration:{{- tool_data['function']['name'] -}}{description:<|\"|>{{- tool_data['function']['description'] -}}<|\"|>\n {%- set params = tool_data['function']['parameters'] -%}\n {%- if params -%}\n ,parameters:{\n {%- if params['properties'] -%}\n properties:{ {{- format_parameters(params['properties'], params['required']) -}} },\n {%- endif -%}\n {%- if params['required'] -%}\n required:[\n {%- for item in params['required'] -%}\n <|\"|>{{- item -}}<|\"|>\n {{- ',' if not loop.last -}}\n {%- endfor -%}\n ],\n {%- endif -%}\n {%- if params['type'] -%}\n type:<|\"|>{{- params['type'] | upper -}}<|\"|>}\n {%- endif -%}\n {%- endif -%}\n {%- if 'response' in tool_data['function'] -%}\n {%- set response_declaration = tool_data['function']['response'] -%}\n ,response:{\n {%- if response_declaration['description'] -%}\n description:<|\"|>{{- response_declaration['description'] -}}<|\"|>,\n {%- endif -%}\n {%- if response_declaration['type'] | upper == 'OBJECT' -%}\n type:<|\"|>{{- response_declaration['type'] | upper -}}<|\"|>}\n {%- endif -%}\n {%- endif -%}\n }\n{%- endmacro -%}\n{%- macro format_argument(argument, escape_keys=True) -%}\n {%- if argument is string -%}\n {{- '<|\"|>' + argument + '<|\"|>' -}}\n {%- elif argument is boolean -%}\n {{- 'true' if argument else 'false' -}}\n {%- elif argument is mapping -%}\n {{- '{' -}}\n {%- set ns = namespace(found_first=false) -%}\n {%- for key, value in argument | dictsort -%}\n {%- if ns.found_first %},{% endif -%}\n {%- set ns.found_first = true -%}\n {%- if escape_keys -%}\n {{- '<|\"|>' + key + '<|\"|>' -}}\n {%- else -%}\n {{- key -}}\n {%- endif -%}\n :{{- format_argument(value, escape_keys=escape_keys) -}}\n {%- endfor -%}\n {{- '}' -}}\n {%- elif argument is sequence -%}\n {{- '[' -}}\n {%- for item in argument -%}\n {{- format_argument(item, escape_keys=escape_keys) -}}\n {%- if not loop.last %},{% endif -%}\n {%- endfor -%}\n {{- ']' -}}\n {%- else -%}\n {{- argument -}}\n {%- endif -%}\n{%- endmacro -%}\n{%- macro strip_thinking(text) -%}\n {%- set ns = namespace(result='') -%}\n {%- for part in text.split('') -%}\n {%- if '<|channel>' in part -%}\n {%- set ns.result = ns.result + part.split('<|channel>')[0] -%}\n {%- else -%}\n {%- set ns.result = ns.result + part -%}\n {%- endif -%}\n {%- endfor -%}\n {{- ns.result | trim -}}\n{%- endmacro -%}\n\n{%- set ns = namespace(prev_message_type=None) -%}\n{%- set loop_messages = messages -%}\n{{ bos_token }}\n{#- Handle System/Tool Definitions Block -#}\n{%- if (enable_thinking is defined and enable_thinking) or tools or messages[0]['role'] in ['system', 'developer'] -%}\n {{- '<|turn>system\\n' -}}\n\n {#- Inject Thinking token at the very top of the FIRST system turn -#}\n {%- if enable_thinking is defined and enable_thinking -%}\n {{- '<|think|>' -}}\n {%- set ns.prev_message_type = 'think' -%}\n {%- endif -%}\n\n {%- if messages[0]['role'] in ['system', 'developer'] -%}\n {{- messages[0]['content'] | trim -}}\n {%- set loop_messages = messages[1:] -%}\n {%- endif -%}\n\n {%- if tools -%}\n {%- for tool in tools %}\n {{- '<|tool>' -}}\n {{- format_function_declaration(tool) | trim -}}\n {{- '' -}}\n {%- endfor %}\n {%- set ns.prev_message_type = 'tool' -%}\n {%- endif -%}\n\n {{- '\\n' -}}\n{%- endif %}\n\n{#- Loop through messages -#}\n{%- for message in loop_messages -%}\n {%- set ns.prev_message_type = None -%}\n {%- set role = 'model' if message['role'] == 'assistant' else message['role'] -%}\n {{- '<|turn>' + role + '\\n' }}\n\n {%- if message['tool_calls'] -%}\n {%- for tool_call in message['tool_calls'] -%}\n {%- set function = tool_call['function'] -%}\n {{- '<|tool_call>call:' + function['name'] + '{' -}}\n {%- if function['arguments'] is mapping -%}\n {%- set ns_args = namespace(found_first=false) -%}\n {%- for key, value in function['arguments'] | dictsort -%}\n {%- if ns_args.found_first %},{% endif -%}\n {%- set ns_args.found_first = true -%}\n {{- key -}}:{{- format_argument(value, escape_keys=False) -}}\n {%- endfor -%}\n {%- elif function['arguments'] is string -%}\n {{- function['arguments'] -}}\n {%- endif -%}\n {{- '}' -}}\n {%- endfor -%}\n {%- set ns.prev_message_type = 'tool_call' -%}\n {%- endif -%}\n\n {%- if message['tool_responses'] -%}\n {#- Tool Response handling -#}\n {%- for tool_response in message['tool_responses'] -%}\n {{- '<|tool_response>' -}}\n {%- if tool_response['response'] is mapping -%}\n {{- 'response:' + tool_response['name'] | default('unknown') + '{' -}}\n {%- for key, value in tool_response['response'] | dictsort -%}\n {{- key -}}:{{- format_argument(value, escape_keys=False) -}}\n {%- if not loop.last %},{% endif -%}\n {%- endfor -%}\n {{- '}' -}}\n {%- else -%}\n {{- 'response:' + tool_response['name'] | default('unknown') + '{value:' + format_argument(tool_response['response'], escape_keys=False) + '}' -}}\n {%- endif -%}\n {{- '' -}}\n {%- endfor -%}\n {%- set ns.prev_message_type = 'tool_response' -%}\n {%- endif -%}\n\n {%- if message['content'] is string -%}\n {%- if role == 'model' -%}\n {{- strip_thinking(message['content']) -}}\n {%- else -%}\n {{- message['content'] | trim -}}\n {%- endif -%}\n {%- elif message['content'] is sequence -%}\n {%- for item in message['content'] -%}\n {%- if item['type'] == 'text' -%}\n {%- if role == 'model' -%}\n {{- strip_thinking(item['text']) -}}\n {%- else -%}\n {{- item['text'] | trim -}}\n {%- endif -%}\n {%- elif item['type'] == 'image' -%}\n {{- '\\n\\n<|image|>\\n\\n' -}}\n {%- set ns.prev_message_type = 'image' -%}\n {%- elif item['type'] == 'audio' -%}\n {{- '<|audio|>' -}}\n {%- set ns.prev_message_type = 'audio' -%}\n {%- elif item['type'] == 'video' -%}\n {{- '\\n\\n<|video|>\\n\\n' -}}\n {%- set ns.prev_message_type = 'video' -%}\n {%- endif -%}\n {%- endfor -%}\n {%- endif -%}\n\n {%- if not (message['tool_responses'] and not message['content']) -%}\n {{- '\\n' -}}\n {%- endif -%}\n{%- endfor -%}\n\n{%- if add_generation_prompt -%}\n {%- if ns.prev_message_type != 'tool_response' -%}\n {{- '<|turn>model\\n' -}}\n {%- endif -%}\n{%- endif -%}", "values_redacted": true, "kv_count": 59 }, "tokenizer": { "type": "gemma4", "vocab_size": 262144, "bos_token_id": 2, "eos_token_id": 1, "has_chat_template": true, "chat_template_chars": 11926 }, "tensors": { "count": 720, "preview": [ { "index": 0, "name": "output_norm.weight", "dims": [ 2560 ], "dtype_name": "F32", "dtype_id": 0, "offset": 0, "byte_len": 10240, "bits_per_elem": 32.0 }, { "index": 1, "name": "per_layer_model_proj.weight", "dims": [ 2560, 10752 ], "dtype_name": "F16", "dtype_id": 1, "offset": 10240, "byte_len": 55050240, "bits_per_elem": 16.0 }, { "index": 2, "name": "per_layer_proj_norm.weight", "dims": [ 256 ], "dtype_name": "F32", "dtype_id": 0, "offset": 55060480, "byte_len": 1024, "bits_per_elem": 32.0 }, { "index": 3, "name": "per_layer_token_embd.weight", "dims": [ 10752, 262144 ], "dtype_name": "Q6_K", "dtype_id": 14, "offset": 55061504, "byte_len": 2312110080, "bits_per_elem": 6.5625 }, { "index": 4, "name": "rope_freqs.weight", "dims": [ 256 ], "dtype_name": "F32", "dtype_id": 0, "offset": 2367171584, "byte_len": 1024, "bits_per_elem": 32.0 }, { "index": 5, "name": "token_embd.weight", "dims": [ 2560, 262144 ], "dtype_name": "Q6_K", "dtype_id": 14, "offset": 2367172608, "byte_len": 550502400, "bits_per_elem": 6.5625 }, { "index": 6, "name": "blk.0.attn_k.weight", "dims": [ 2560, 512 ], "dtype_name": "IQ3_S", "dtype_id": 21, "offset": 2917675008, "byte_len": 563200, "bits_per_elem": 3.4375 }, { "index": 7, "name": "blk.0.attn_k_norm.weight", "dims": [ 256 ], "dtype_name": "F32", "dtype_id": 0, "offset": 2918238208, "byte_len": 1024, "bits_per_elem": 32.0 }, { "index": 8, "name": "blk.0.attn_norm.weight", "dims": [ 2560 ], "dtype_name": "F32", "dtype_id": 0, "offset": 2918239232, "byte_len": 10240, "bits_per_elem": 32.0 }, { "index": 9, "name": "blk.0.attn_output.weight", "dims": [ 2048, 2560 ], "dtype_name": "Q4_K", "dtype_id": 12, "offset": 2918249472, "byte_len": 2949120, "bits_per_elem": 4.5 }, { "index": 10, "name": "blk.0.attn_q.weight", "dims": [ 2560, 2048 ], "dtype_name": "IQ3_S", "dtype_id": 21, "offset": 2921198592, "byte_len": 2252800, "bits_per_elem": 3.4375 }, { "index": 11, "name": "blk.0.attn_q_norm.weight", "dims": [ 256 ], "dtype_name": "F32", "dtype_id": 0, "offset": 2923451392, "byte_len": 1024, "bits_per_elem": 32.0 }, { "index": 12, "name": "blk.0.attn_v.weight", "dims": [ 2560, 512 ], "dtype_name": "Q4_K", "dtype_id": 12, "offset": 2923452416, "byte_len": 737280, "bits_per_elem": 4.5 }, { "index": 13, "name": "blk.0.ffn_down.weight", "dims": [ 10240, 2560 ], "dtype_name": "Q4_K", "dtype_id": 12, "offset": 2924189696, "byte_len": 14745600, "bits_per_elem": 4.5 }, { "index": 14, "name": "blk.0.ffn_gate.weight", "dims": [ 2560, 10240 ], "dtype_name": "IQ3_S", "dtype_id": 21, "offset": 2938935296, "byte_len": 11264000, "bits_per_elem": 3.4375 }, { "index": 15, "name": "blk.0.ffn_norm.weight", "dims": [ 2560 ], "dtype_name": "F32", "dtype_id": 0, "offset": 2950199296, "byte_len": 10240, "bits_per_elem": 32.0 }, { "index": 16, "name": "blk.0.ffn_up.weight", "dims": [ 2560, 10240 ], "dtype_name": "IQ3_S", "dtype_id": 21, "offset": 2950209536, "byte_len": 11264000, "bits_per_elem": 3.4375 }, { "index": 17, "name": "blk.0.inp_gate.weight", "dims": [ 2560, 256 ], "dtype_name": "IQ3_S", "dtype_id": 21, "offset": 2961473536, "byte_len": 281600, "bits_per_elem": 3.4375 }, { "index": 18, "name": "blk.0.layer_output_scale.weight", "dims": [ 1 ], "dtype_name": "F32", "dtype_id": 0, "offset": 2961755136, "byte_len": 32, "bits_per_elem": 256.0 }, { "index": 19, "name": "blk.0.post_attention_norm.weight", "dims": [ 2560 ], "dtype_name": "F32", "dtype_id": 0, "offset": 2961755168, "byte_len": 10240, "bits_per_elem": 32.0 }, { "index": 20, "name": "blk.0.post_ffw_norm.weight", "dims": [ 2560 ], "dtype_name": "F32", "dtype_id": 0, "offset": 2961765408, "byte_len": 10240, "bits_per_elem": 32.0 }, { "index": 21, "name": "blk.0.post_norm.weight", "dims": [ 2560 ], "dtype_name": "F32", "dtype_id": 0, "offset": 2961775648, "byte_len": 10240, "bits_per_elem": 32.0 }, { "index": 22, "name": "blk.0.proj.weight", "dims": [ 256, 2560 ], "dtype_name": "IQ3_S", "dtype_id": 21, "offset": 2961785888, "byte_len": 281600, "bits_per_elem": 3.4375 }, { "index": 23, "name": "blk.1.attn_k.weight", "dims": [ 2560, 512 ], "dtype_name": "IQ3_S", "dtype_id": 21, "offset": 2962067488, "byte_len": 563200, "bits_per_elem": 3.4375 }, { "index": 24, "name": "blk.1.attn_k_norm.weight", "dims": [ 256 ], "dtype_name": "F32", "dtype_id": 0, "offset": 2962630688, "byte_len": 1024, "bits_per_elem": 32.0 }, { "index": 25, "name": "blk.1.attn_norm.weight", "dims": [ 2560 ], "dtype_name": "F32", "dtype_id": 0, "offset": 2962631712, "byte_len": 10240, "bits_per_elem": 32.0 }, { "index": 26, "name": "blk.1.attn_output.weight", "dims": [ 2048, 2560 ], "dtype_name": "Q4_K", "dtype_id": 12, "offset": 2962641952, "byte_len": 2949120, "bits_per_elem": 4.5 }, { "index": 27, "name": "blk.1.attn_q.weight", "dims": [ 2560, 2048 ], "dtype_name": "IQ3_S", "dtype_id": 21, "offset": 2965591072, "byte_len": 2252800, "bits_per_elem": 3.4375 }, { "index": 28, "name": "blk.1.attn_q_norm.weight", "dims": [ 256 ], "dtype_name": "F32", "dtype_id": 0, "offset": 2967843872, "byte_len": 1024, "bits_per_elem": 32.0 }, { "index": 29, "name": "blk.1.attn_v.weight", "dims": [ 2560, 512 ], "dtype_name": "Q4_K", "dtype_id": 12, "offset": 2967844896, "byte_len": 737280, "bits_per_elem": 4.5 }, { "index": 30, "name": "blk.1.ffn_down.weight", "dims": [ 10240, 2560 ], "dtype_name": "Q4_K", "dtype_id": 12, "offset": 2968582176, "byte_len": 14745600, "bits_per_elem": 4.5 }, { "index": 31, "name": "blk.1.ffn_gate.weight", "dims": [ 2560, 10240 ], "dtype_name": "IQ3_S", "dtype_id": 21, "offset": 2983327776, "byte_len": 11264000, "bits_per_elem": 3.4375 }, { "index": 32, "name": "blk.1.ffn_norm.weight", "dims": [ 2560 ], "dtype_name": "F32", "dtype_id": 0, "offset": 2994591776, "byte_len": 10240, "bits_per_elem": 32.0 }, { "index": 33, "name": "blk.1.ffn_up.weight", "dims": [ 2560, 10240 ], "dtype_name": "IQ3_S", "dtype_id": 21, "offset": 2994602016, "byte_len": 11264000, "bits_per_elem": 3.4375 }, { "index": 34, "name": "blk.1.inp_gate.weight", "dims": [ 2560, 256 ], "dtype_name": "IQ3_S", "dtype_id": 21, "offset": 3005866016, "byte_len": 281600, "bits_per_elem": 3.4375 }, { "index": 35, "name": "blk.1.layer_output_scale.weight", "dims": [ 1 ], "dtype_name": "F32", "dtype_id": 0, "offset": 3006147616, "byte_len": 32, "bits_per_elem": 256.0 }, { "index": 36, "name": "blk.1.post_attention_norm.weight", "dims": [ 2560 ], "dtype_name": "F32", "dtype_id": 0, "offset": 3006147648, "byte_len": 10240, "bits_per_elem": 32.0 }, { "index": 37, "name": "blk.1.post_ffw_norm.weight", "dims": [ 2560 ], "dtype_name": "F32", "dtype_id": 0, "offset": 3006157888, "byte_len": 10240, "bits_per_elem": 32.0 }, { "index": 38, "name": "blk.1.post_norm.weight", "dims": [ 2560 ], "dtype_name": "F32", "dtype_id": 0, "offset": 3006168128, "byte_len": 10240, "bits_per_elem": 32.0 }, { "index": 39, "name": "blk.1.proj.weight", "dims": [ 256, 2560 ], "dtype_name": "IQ3_S", "dtype_id": 21, "offset": 3006178368, "byte_len": 281600, "bits_per_elem": 3.4375 }, { "index": 40, "name": "blk.2.attn_k.weight", "dims": [ 2560, 512 ], "dtype_name": "IQ3_S", "dtype_id": 21, "offset": 3006459968, "byte_len": 563200, "bits_per_elem": 3.4375 }, { "index": 41, "name": "blk.2.attn_k_norm.weight", "dims": [ 256 ], "dtype_name": "F32", "dtype_id": 0, "offset": 3007023168, "byte_len": 1024, "bits_per_elem": 32.0 }, { "index": 42, "name": "blk.2.attn_norm.weight", "dims": [ 2560 ], "dtype_name": "F32", "dtype_id": 0, "offset": 3007024192, "byte_len": 10240, "bits_per_elem": 32.0 }, { "index": 43, "name": "blk.2.attn_output.weight", "dims": [ 2048, 2560 ], "dtype_name": "Q4_K", "dtype_id": 12, "offset": 3007034432, "byte_len": 2949120, "bits_per_elem": 4.5 }, { "index": 44, "name": "blk.2.attn_q.weight", "dims": [ 2560, 2048 ], "dtype_name": "IQ3_S", "dtype_id": 21, "offset": 3009983552, "byte_len": 2252800, "bits_per_elem": 3.4375 }, { "index": 45, "name": "blk.2.attn_q_norm.weight", "dims": [ 256 ], "dtype_name": "F32", "dtype_id": 0, "offset": 3012236352, "byte_len": 1024, "bits_per_elem": 32.0 }, { "index": 46, "name": "blk.2.attn_v.weight", "dims": [ 2560, 512 ], "dtype_name": "Q4_K", "dtype_id": 12, "offset": 3012237376, "byte_len": 737280, "bits_per_elem": 4.5 }, { "index": 47, "name": "blk.2.ffn_down.weight", "dims": [ 10240, 2560 ], "dtype_name": "Q4_K", "dtype_id": 12, "offset": 3012974656, "byte_len": 14745600, "bits_per_elem": 4.5 }, { "index": 48, "name": "blk.2.ffn_gate.weight", "dims": [ 2560, 10240 ], "dtype_name": "IQ3_S", "dtype_id": 21, "offset": 3027720256, "byte_len": 11264000, "bits_per_elem": 3.4375 }, { "index": 49, "name": "blk.2.ffn_norm.weight", "dims": [ 2560 ], "dtype_name": "F32", "dtype_id": 0, "offset": 3038984256, "byte_len": 10240, "bits_per_elem": 32.0 } ], "dtype_histogram": { "F32": 339, "F16": 1, "Q6_K": 2, "IQ3_S": 289, "Q4_K": 89 }, "dtype_stats": { "F32": { "count": 339, "bytes": 2264384, "mb": 2.16, "avg_bits_per_elem": 32.016628 }, "F16": { "count": 1, "bytes": 55050240, "mb": 52.5, "avg_bits_per_elem": 16.0 }, "Q6_K": { "count": 2, "bytes": 2862612480, "mb": 2730.0, "avg_bits_per_elem": 6.5625 }, "IQ3_S": { "count": 289, "bytes": 1524582400, "mb": 1453.96, "avg_bits_per_elem": 3.4375 }, "Q4_K": { "count": 89, "bytes": 254361600, "mb": 242.58, "avg_bits_per_elem": 4.5 } }, "integrity": { "offsets_monotone": true, "overlapping_tensor_pairs": 0, "misaligned_tensors": 0, "oob_tensor_count": 0, "tensor_bytes_sum": 4698871104, "tensor_delta_bytes": 15820160, "tensor_delta_mb": 15.09, "header_kv_dir_bytes": 15820160, "trailing_bytes": 0, "unaccounted_bytes": 0 } }, "security": { "grade": "B", "score": 81, "findings": [ { "severity": "INFO", "category": "Architecture", "title": "Grouped Query Attention (GQA) detected", "detail": "head_count=8, head_count_kv=2 — requires GQA-aware runtime", "recommendation": "Ensure your runtime (llama.cpp >= b1500, Ollama >= 0.1.30) supports GQA" }, { "severity": "MEDIUM", "category": "Metadata Security", "title": "Complex agentic chat template — context-dependent risk", "detail": "Template uses tool_calls, tool responses, namespace() — the full pattern for agentic pipelines. These are legitimate HuggingFace constructs but represent the highest-complexity template class. Risk is context-dependent: negligible for static inference, meaningful in autonomous-agent deployments where user-controlled content can reach the template renderer.", "recommendation": "Review template before use in agentic pipelines. Use sandboxed Jinja2 rendering and compare against the upstream author's reference template." }, { "severity": "MEDIUM", "category": "Metadata Security", "title": "External URLs embedded in 3 KV metadata field(s)", "detail": "Fields: ['general.base_model.0.repo_url', 'general.url', 'general.source.url']", "recommendation": "URLs in model metadata may be used for tracking or exfiltration when metadata is rendered; do not auto-fetch" }, { "severity": "INFO", "category": "Embedding Size", "title": "Large embedding tensor 'per_layer_token_embd.weight': 2,312,110,080 bytes", "detail": "vocab=262,144 x emb_dim=10,752 x 0.8203125 B/elem -> expected 2,312,110,080 B — size is consistent with architecture", "recommendation": "No action needed; large embeddings are expected for large-vocabulary models" }, { "severity": "INFO", "category": "Embedding Size", "title": "Large embedding tensor 'token_embd.weight': 550,502,400 bytes", "detail": "vocab=262,144 x emb_dim=2,560 x 0.8203125 B/elem -> expected 550,502,400 B — size is consistent with architecture", "recommendation": "No action needed; large embeddings are expected for large-vocabulary models" }, { "severity": "LOW", "category": "Statistical Anomaly", "title": "134 weight tensor(s) are >20x the median size: per_layer_model_proj.weight, blk.0.ffn_down.weight, blk.0.ffn_gate.weight", "detail": "Extreme outlier tensors may contain embedded payloads or crafted weight matrices", "recommendation": "Inspect outlier tensor names and compare against reference architecture" }, { "severity": "INFO", "category": "Quantization", "title": "Model is quantized: IQ3_S, Q4_K, Q6_K", "detail": "Quantization reduces size and inference cost at the expense of some accuracy", "recommendation": "Verify the quantization level is appropriate for your use case (Q4_K_M is a common balanced choice)" }, { "severity": "INFO", "category": "Quantization", "title": "Declared file_type (IQ3_S) does not match effective tensor dtypes (Q6_K)", "detail": "The GGUF general.file_type field declares 'IQ3_S' but the actual tensors in this file are 'Q6_K'. This is common for multi-file model packages where the projector/adapter GGUF inherits the main model's file_type metadata. The effective dtype shown in the report reflects actual tensor storage.", "recommendation": "No action required if this is a companion file (mmproj, adapter). For standalone models, this may indicate incorrect GGUF export." } ] }, "compatibility": [ { "runtime": "llama.cpp", "supported": null, "container": true, "quant": true, "arch": null, "context": "128k — may require flash attention (--flash-attn) and large VRAM", "notes": "GGUF v3 container: supported | quant Q6_K: supported | 128k context: requires sufficient VRAM/RAM and --ctx-size flag — architecture 'gemma4' is uncommon; check runtime release notes for explicit support" }, { "runtime": "Ollama", "supported": null, "container": true, "quant": true, "arch": null, "context": "128k — may require flash attention (--flash-attn) and large VRAM", "notes": "GGUF via llama.cpp backend: container and quantization supported — architecture 'gemma4' is uncommon; check runtime release notes for explicit support" }, { "runtime": "LM Studio", "supported": null, "container": true, "quant": true, "arch": null, "context": "128k — may require flash attention (--flash-attn) and large VRAM", "notes": "GGUF container and quantization supported; GPU/CPU inference — architecture 'gemma4' is uncommon; check runtime release notes for explicit support" }, { "runtime": "vLLM", "supported": false, "container": false, "quant": false, "arch": null, "context": null, "not_supported_msg": "GGUF not supported — requires safetensors format", "notes": "GGUF not supported; requires safetensors format" }, { "runtime": "HuggingFace Transformers", "container": false, "quant": false, "arch": null, "context": null, "supported": false, "not_supported_msg": "GGUF not directly supported — use safetensors or PyTorch checkpoint", "notes": "GGUF not supported; requires safetensors or PyTorch checkpoint" } ], "report": { "summary": "GGUF v3 model | architecture: gemma4 | size: 7.5B | quantization: Q6_K | 720 tensors | 3 warnings / 5 informational", "bullets": [ "GGUF v3 model | architecture: gemma4 | size: 7.5B | quantization: Q6_K | 720 tensors | 3 warnings / 5 informational", "Size: 4496.28 MB", "SHA256: 434ed4a393864978..." ], "recommendations": [ "Review template before use in agentic pipelines. Use sandboxed Jinja2 rendering and compare against the upstream author's reference template.", "URLs in model metadata may be used for tracking or exfiltration when metadata is rendered; do not auto-fetch", "Inspect outlier tensor names and compare against reference architecture" ] } }