Upload folder using huggingface_hub

.gitattributes

@@ -1,35 +1,4 @@
-*.7z filter=lfs diff=lfs merge=lfs -text
-*.arrow filter=lfs diff=lfs merge=lfs -text
+*.rows filter=lfs diff=lfs merge=lfs -text
 *.bin filter=lfs diff=lfs merge=lfs -text
-*.bz2 filter=lfs diff=lfs merge=lfs -text
-*.ckpt filter=lfs diff=lfs merge=lfs -text
-*.ftz filter=lfs diff=lfs merge=lfs -text
-*.gz filter=lfs diff=lfs merge=lfs -text
-*.h5 filter=lfs diff=lfs merge=lfs -text
-*.joblib filter=lfs diff=lfs merge=lfs -text
-*.lfs.* filter=lfs diff=lfs merge=lfs -text
-*.mlmodel filter=lfs diff=lfs merge=lfs -text
-*.model filter=lfs diff=lfs merge=lfs -text
-*.msgpack filter=lfs diff=lfs merge=lfs -text
-*.npy filter=lfs diff=lfs merge=lfs -text
-*.npz filter=lfs diff=lfs merge=lfs -text
-*.onnx filter=lfs diff=lfs merge=lfs -text
-*.ot filter=lfs diff=lfs merge=lfs -text
-*.parquet filter=lfs diff=lfs merge=lfs -text
-*.pb filter=lfs diff=lfs merge=lfs -text
-*.pickle filter=lfs diff=lfs merge=lfs -text
-*.pkl filter=lfs diff=lfs merge=lfs -text
-*.pt filter=lfs diff=lfs merge=lfs -text
-*.pth filter=lfs diff=lfs merge=lfs -text
-*.rar filter=lfs diff=lfs merge=lfs -text
-*.safetensors filter=lfs diff=lfs merge=lfs -text
-saved_model/**/* filter=lfs diff=lfs merge=lfs -text
-*.tar.* filter=lfs diff=lfs merge=lfs -text
-*.tar filter=lfs diff=lfs merge=lfs -text
-*.tflite filter=lfs diff=lfs merge=lfs -text
-*.tgz filter=lfs diff=lfs merge=lfs -text
-*.wasm filter=lfs diff=lfs merge=lfs -text
-*.xz filter=lfs diff=lfs merge=lfs -text
-*.zip filter=lfs diff=lfs merge=lfs -text
-*.zst filter=lfs diff=lfs merge=lfs -text
-*tfevents* filter=lfs diff=lfs merge=lfs -text
+model.safetensors filter=lfs diff=lfs merge=lfs -text
+tokenizer.json filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,46 @@
+---
+language:
+- en
+- zh
+license: mit
+base_model: pipenetwork/GLM-5.3-Flash-MLX-4bit
+library_name: gbx-lm
+pipeline_tag: image-text-to-text
+tags:
+- mlx
+- apple-silicon
+- expert-paging
+- mixture-of-experts
+- multimodal
+---
+
+# GLM-5.3-Flash-4bit-paged
+
+Expert-paged build of `pipenetwork/GLM-5.3-Flash-MLX-4bit`. The weights that are read a fraction at a time
+live in their own containers, so a machine loads what it needs rather than all
+of it.
+
+| file | size | holds |
+| --- | --- | --- |
+| `model.safetensors` | 5.89 GiB | resident weights |
+| `experts.bin` | 159.47 GiB | routed experts |
+
+Total 165.37 GiB — the same as the source. The bytes moved, they were not copied.
+
+```python
+from gbx_lm.utils import load
+model, tokenizer = load("GreenBitAI/GLM-5.3-Flash-4bit-paged")
+```
+
+Where the weights fit they are filled from `experts.bin` and the model runs the
+stock path at stock speed; where they do not, they stream from disk. Reading
+the machine decides that, not a flag.
+
+To override that: `GBX_PAGING=off` holds the experts resident.
+
+Checked at build time, while the source checkpoint was still there to compare against:
+
+- **SKIP** bit-identical logits — cannot hold a resident reference: 165.4 GiB checkpoint needs ~198.4 GiB resident, 181.8 GiB usable
+- **PASS** layer-wise vs resident — 42 layers x 2 draws exact, 60.93 GiB peak for this gate
+
+Quantization, tokenizer, chat template and licence are unchanged from `pipenetwork/GLM-5.3-Flash-MLX-4bit`.

chat_template.jinja

@@ -0,0 +1,257 @@
+[gMASK]<sop>
+{%- set effective_reasoning_effort = reasoning_effort if reasoning_effort is defined and reasoning_effort in ['low', 'high'] else 'max' -%}
+{%- if effective_reasoning_effort is not none -%}<|system|>Reasoning Effort: {{ effective_reasoning_effort | capitalize }}{%- endif -%}
+{%- set clear_thinking = clear_thinking if clear_thinking is defined else false -%}
+{%- if tools -%}
+{%- macro tool_to_json(tool) -%}
+    {%- set ns_tool = namespace(first=true) -%}
+    {{ '{' -}}
+    {%- for k, v in tool.items() -%}
+        {%- if k != 'defer_loading' and k != 'strict' -%}
+            {%- if not ns_tool.first -%}{{- ', ' -}}{%- endif -%}
+            {%- set ns_tool.first = false -%}
+            "{{ k }}": {{ v | tojson(ensure_ascii=False) }}
+        {%- endif -%}
+    {%- endfor -%}
+    {{- '}' -}}
+{%- endmacro -%}
+{%- macro tool_references_to_response(refs) -%}
+    {{- '<tool_response><tools>\n' -}}
+    {%- for tr in refs -%}
+        {%- for tool in tools -%}
+            {%- if 'function' in tool -%}
+                {%- set tool = tool['function'] -%}
+            {%- endif -%}
+            {%- if tool.name == tr.name -%}
+                {{- tool_to_json(tool) + '\n' -}}
+            {%- endif -%}
+        {%- endfor -%}
+    {%- endfor -%}
+    {{- '</tools></tool_response>' -}}
+{%- endmacro -%}
+<|system|>
+# Tools
+
+You may call one or more functions to assist with the user query.
+
+You are provided with function signatures within <tools></tools> XML tags:
+<tools>
+{% for tool in tools %}
+{%- if 'function' in tool -%}
+    {%- set tool = tool['function'] -%}
+{%- endif -%}
+{% if tool.defer_loading is not defined or not tool.defer_loading %}
+{{ tool_to_json(tool) }}
+{% endif %}
+{% endfor %}
+</tools>
+
+For each function call, output the function name and arguments within the following XML format:
+<tool_call>{function-name}<arg_key>{arg-key-1}</arg_key><arg_value>{arg-value-1}</arg_value><arg_key>{arg-key-2}</arg_key><arg_value>{arg-value-2}</arg_value>...</tool_call>{%- endif -%}
+{%- macro emit_image() -%}<|begin_of_image|><|image|><|end_of_image|>{%- endmacro -%}
+{%- macro emit_video() -%}<|begin_of_video|><|video|><|end_of_video|>{%- endmacro -%}
+{%- macro emit_audio() -%}<|begin_of_audio|><|end_of_audio|>{%- endmacro -%}
+{%- macro visible_text(content) -%}
+    {%- if content is string -%}
+        {{- content -}}
+    {%- elif content is iterable and content is not mapping -%}
+        {%- for item in content -%}
+            {%- if item is mapping and item.type == 'text' -%}
+                {{- item.text -}}
+            {%- elif item is string -%}
+                {{- item -}}
+            {%- elif item is mapping and item.type in ['image', 'image_url'] -%}
+                {{- emit_image() -}}
+            {%- elif item is mapping and item.type in ['video', 'video_url'] -%}
+                {{- emit_video() -}}
+            {%- elif item is mapping and item.type in ['audio', 'audio_url', 'input_audio'] -%}
+                {{- emit_audio() -}}
+            {%- endif -%}
+        {%- endfor -%}
+    {%- else -%}
+        {{- content }}
+    {%- endif -%}
+{%- endmacro -%}
+{%- macro tool_response(text) -%}
+{{- '<tool_response>' + text + '</tool_response>' -}}
+{%- endmacro -%}
+{%- macro render_tool_response(m) -%}
+{%- if m.content is string -%}
+    {{- tool_response(m.content) -}}
+{%- elif m.content and m.content is not mapping and m.content.0.type == "tool_reference" -%}
+    {{- tool_references_to_response(m.content) -}}
+{%- elif is_list_of_outputs(m) -%}
+    {%- for tr in m.content -%}
+        {%- if tr.output is iterable and tr.output is not string and tr.output is not mapping and tr.output and tr.output.0.type == "tool_reference" -%}
+            {{- tool_references_to_response(tr.output) -}}
+        {%- else -%}
+            {{- tool_response(visible_text(tr.output)) -}}
+        {%- endif -%}
+    {%- endfor -%}
+{%- else -%}
+    {{- tool_response(visible_text(m.content)) -}}
+{%- endif -%}
+{%- endmacro -%}
+{%- macro id_of(obj) -%}
+    {%- if obj.tool_call_id -%}
+        {{- obj.tool_call_id -}}
+    {%- elif obj.id -%}
+        {{- obj.id -}}
+    {%- endif -%}
+{%- endmacro -%}
+{%- macro is_list_of_outputs(m) -%}
+    {%- if m.content and m.content.0.output is defined -%}1{%- endif -%}
+{%- endmacro -%}
+{%- macro has_dup_tool_result_id(lo, hi, target) -%}
+    {%- set ns_cnt = namespace(n=0) -%}
+    {%- for k in range(lo, hi + 1) -%}
+        {%- set m = messages[k] -%}
+        {%- if is_list_of_outputs(m) -%}
+            {%- for entry in m.content -%}
+                {%- if id_of(entry) == target -%}
+                    {%- set ns_cnt.n = ns_cnt.n + 1 -%}
+                {%- endif -%}
+            {%- endfor -%}
+        {%- elif id_of(m) == target -%}
+            {%- set ns_cnt.n = ns_cnt.n + 1 -%}
+        {%- endif -%}
+        {%- if ns_cnt.n > 1 -%}{%- break -%}{%- endif -%}
+    {%- endfor -%}
+    {%- if ns_cnt.n > 1 -%}1{%- endif -%}
+{%- endmacro -%}
+{%- macro tc_id_exists(tcs, target) -%}
+    {%- set ns_f = namespace(found=false) -%}
+    {%- for tc in tcs -%}
+        {%- if id_of(tc) == target -%}
+            {%- set ns_f.found = true -%}
+            {%- break -%}
+        {%- endif -%}
+    {%- endfor -%}
+    {%- if ns_f.found -%}1{%- endif -%}
+{%- endmacro -%}
+{%- set ns = namespace(last_user_index=-1) -%}
+{%- for m in messages %}
+    {%- if m.role == 'user' %}
+        {%- set ns.last_user_index = loop.index0 -%}
+    {%- endif %}
+{%- endfor %}
+{%- for m in messages -%}
+{%- if m.role == 'user' -%}<|user|>{{ visible_text(m.content) }}
+{%- elif m.role == 'assistant' -%}
+<|assistant|>
+{%- set content = visible_text(m.content) %}
+{%- if m.reasoning_content is string %}
+    {%- set reasoning_content = m.reasoning_content %}
+{%- elif '</think>' in content %}
+    {%- set reasoning_content = content.split('</think>')[0].split('<think>')[-1] %}
+    {%- set content = content.split('</think>')[-1] %}
+{%- endif %}
+{%- if (not clear_thinking or loop.index0 > ns.last_user_index) and reasoning_content is defined -%}
+{{ '<think>' + reasoning_content +  '</think>'}}
+{%- else -%}
+{{ '<think></think>' }}
+{%- endif -%}
+{%- if content.strip() -%}
+{{ content.strip() }}
+{%- endif -%}
+{% if m.tool_calls %}
+{% for tc in m.tool_calls %}
+{%- if tc.function %}
+    {%- set tc = tc.function %}
+{%- endif %}
+{{- '<tool_call>' + tc.name -}}
+{% set _args = tc.arguments %}{% for k, v in _args.items() %}<arg_key>{{ k }}</arg_key><arg_value>{{ v | tojson(ensure_ascii=False) if v is not string else v }}</arg_value>{% endfor %}</tool_call>{% endfor %}
+{% endif %}
+{%- elif m.role == 'tool' -%}
+{%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}
+    {{- '<|observation|>' -}}
+    {%- set block_start = loop.index0 -%}
+    {%- set ns_blk = namespace(end=block_start) -%}
+    {%- for j in range(block_start, messages|length) -%}
+        {%- if messages[j].role == 'tool' -%}
+            {%- set ns_blk.end = j -%}
+        {%- else -%}
+            {%- break -%}
+        {%- endif -%}
+    {%- endfor -%}
+    {%- set ns_a = namespace(tool_calls=none) -%}
+    {%- if block_start > 0 and messages[block_start - 1].role == 'assistant' and messages[block_start - 1].tool_calls -%}
+        {%- set ns_a.tool_calls = messages[block_start - 1].tool_calls -%}
+    {%- endif -%}
+    {%- set ns_chk = namespace(can_sort=true) -%}
+    {%- if not ns_a.tool_calls -%}
+        {%- set ns_chk.can_sort = false -%}
+    {%- else -%}
+        {%- for k in range(block_start, ns_blk.end + 1) -%}
+            {%- set m = messages[k] -%}
+            {%- if is_list_of_outputs(m) -%}
+                {%- for entry in m.content -%}
+                    {%- set eid = id_of(entry) -%}
+                    {%- if not eid -%}
+                        {%- set ns_chk.can_sort = false -%}
+                    {%- elif has_dup_tool_result_id(block_start, ns_blk.end, eid) -%}
+                        {%- set ns_chk.can_sort = false -%}
+                    {%- elif not tc_id_exists(ns_a.tool_calls, eid) -%}
+                        {%- set ns_chk.can_sort = false -%}
+                    {%- endif -%}
+                {%- endfor -%}
+            {%- else -%}
+                {%- set tk_id = id_of(m) -%}
+                {%- if not tk_id -%}
+                    {%- set ns_chk.can_sort = false -%}
+                {%- elif has_dup_tool_result_id(block_start, ns_blk.end, tk_id) -%}
+                    {%- set ns_chk.can_sort = false -%}
+                {%- elif not tc_id_exists(ns_a.tool_calls, tk_id) -%}
+                    {%- set ns_chk.can_sort = false -%}
+                {%- endif -%}
+            {%- endif -%}
+        {%- endfor -%}
+        {%- for i in range(ns_a.tool_calls | length) -%}
+            {%- set tc_id = id_of(ns_a.tool_calls[i]) -%}
+            {%- if not tc_id -%}
+                {%- set ns_chk.can_sort = false -%}
+            {%- endif -%}
+            {%- for j in range(i + 1, ns_a.tool_calls | length) -%}
+                {%- if id_of(ns_a.tool_calls[j]) == tc_id -%}
+                    {%- set ns_chk.can_sort = false -%}
+                {%- endif -%}
+            {%- endfor -%}
+    {%- endfor -%}
+    {%- endif -%}
+    {%- if ns_chk.can_sort -%}
+        {%- for tc in ns_a.tool_calls -%}
+            {%- set tc_id = id_of(tc) -%}
+            {%- for k in range(block_start, ns_blk.end + 1) -%}
+                {%- set m = messages[k] -%}
+                {%- if is_list_of_outputs(m) -%}
+                    {%- for entry in m.content -%}
+                        {%- set eid = id_of(entry) -%}
+                        {%- if eid == tc_id -%}
+                            {%- if entry.output is iterable and entry.output is not string and entry.output is not mapping and entry.output and entry.output.0.type == "tool_reference" -%}
+                                {{- tool_references_to_response(entry.output) -}}
+                            {%- else -%}
+                                {{- tool_response(visible_text(entry.output)) -}}
+                            {%- endif -%}
+                        {%- endif -%}
+    {%- endfor -%}
+{%- else -%}
+                    {%- set tk_id = id_of(m) -%}
+                    {%- if tk_id == tc_id -%}
+                        {{- render_tool_response(m) -}}
+                    {%- endif -%}
+                {%- endif -%}
+            {%- endfor -%}
+        {%- endfor -%}
+    {%- else -%}
+        {%- for k in range(block_start, ns_blk.end + 1) -%}
+            {{- render_tool_response(messages[k]) -}}
+        {%- endfor -%}
+    {%- endif -%}
+{% endif -%}
+{%- elif m.role == 'system' -%}
+<|system|>{{ visible_text(m.content) }}
+{%- endif -%}
+{%- endfor -%}
+{%- if add_generation_prompt -%}
+    <|assistant|>{{- '<think>' -}}
+{%- endif -%}

config.json

@@ -0,0 +1,1067 @@
+{
+  "architectures": [
+    "Glm5NextForConditionalGeneration"
+  ],
+  "text_config": {
+    "attention_bias": false,
+    "attention_dropout": 0.0,
+    "dtype": "bfloat16",
+    "eos_token_id": [
+      154820,
+      154827,
+      154829
+    ],
+    "first_k_dense_replace": 3,
+    "hc_eps": 1e-06,
+    "hc_mult": 4,
+    "hc_sinkhorn_iters": 20,
+    "head_dim": 0,
+    "hidden_act": "silu",
+    "hidden_size": 4096,
+    "index_head_dim": 128,
+    "index_kpool": 4,
+    "index_kpool_always_select_tail": true,
+    "index_kpool_compress": true,
+    "index_n_heads": 32,
+    "index_topk": 2048,
+    "index_share_for_mtp_iteration": true,
+    "indexer_rope_interleave": true,
+    "indexer_types": [
+      "full",
+      "full",
+      "full",
+      "full",
+      "full",
+      "full",
+      "full",
+      "full",
+      "full",
+      "full",
+      "full",
+      "full",
+      "full",
+      "full",
+      "full",
+      "full",
+      "full",
+      "full",
+      "full",
+      "full",
+      "full",
+      "full",
+      "full",
+      "full",
+      "full",
+      "full",
+      "full",
+      "full",
+      "full",
+      "full",
+      "full",
+      "full",
+      "full",
+      "full",
+      "full",
+      "full",
+      "full",
+      "full",
+      "full",
+      "full",
+      "full",
+      "full",
+      "full",
+      "full",
+      "full"
+    ],
+    "initializer_range": 0.02,
+    "intermediate_size": 12288,
+    "kv_lora_rank": 512,
+    "layer_types": [
+      "linear_attention",
+      "linear_attention",
+      "linear_attention",
+      "deepseek_sparse_attention",
+      "linear_attention",
+      "linear_attention",
+      "linear_attention",
+      "deepseek_sparse_attention",
+      "linear_attention",
+      "linear_attention",
+      "linear_attention",
+      "deepseek_sparse_attention",
+      "linear_attention",
+      "linear_attention",
+      "linear_attention",
+      "deepseek_sparse_attention",
+      "linear_attention",
+      "linear_attention",
+      "linear_attention",
+      "deepseek_sparse_attention",
+      "linear_attention",
+      "linear_attention",
+      "linear_attention",
+      "deepseek_sparse_attention",
+      "linear_attention",
+      "linear_attention",
+      "linear_attention",
+      "deepseek_sparse_attention",
+      "linear_attention",
+      "linear_attention",
+      "linear_attention",
+      "deepseek_sparse_attention",
+      "linear_attention",
+      "linear_attention",
+      "linear_attention",
+      "deepseek_sparse_attention",
+      "linear_attention",
+      "linear_attention",
+      "linear_attention",
+      "deepseek_sparse_attention",
+      "linear_attention",
+      "linear_attention",
+      "linear_attention",
+      "deepseek_sparse_attention",
+      "linear_attention"
+    ],
+    "linear_attn_config": {
+      "num_heads": 64,
+      "gate_lower_bound": -5.0,
+      "head_dim": 128,
+      "short_conv_kernel_size": 4,
+      "kda_layers": [
+        0,
+        1,
+        2,
+        4,
+        5,
+        6,
+        8,
+        9,
+        10,
+        12,
+        13,
+        14,
+        16,
+        17,
+        18,
+        20,
+        21,
+        22,
+        24,
+        25,
+        26,
+        28,
+        29,
+        30,
+        32,
+        33,
+        34,
+        36,
+        37,
+        38,
+        40,
+        41,
+        42,
+        44
+      ],
+      "full_attn_layers": [
+        3,
+        7,
+        11,
+        15,
+        19,
+        23,
+        27,
+        31,
+        35,
+        39,
+        43
+      ]
+    },
+    "max_position_embeddings": 1048576,
+    "mhc": true,
+    "mla_use_nope": true,
+    "mlp_layer_types": [
+      "dense",
+      "dense",
+      "dense",
+      "sparse",
+      "sparse",
+      "sparse",
+      "sparse",
+      "sparse",
+      "sparse",
+      "sparse",
+      "sparse",
+      "sparse",
+      "sparse",
+      "sparse",
+      "sparse",
+      "sparse",
+      "sparse",
+      "sparse",
+      "sparse",
+      "sparse",
+      "sparse",
+      "sparse",
+      "sparse",
+      "sparse",
+      "sparse",
+      "sparse",
+      "sparse",
+      "sparse",
+      "sparse",
+      "sparse",
+      "sparse",
+      "sparse",
+      "sparse",
+      "sparse",
+      "sparse",
+      "sparse",
+      "sparse",
+      "sparse",
+      "sparse",
+      "sparse",
+      "sparse",
+      "sparse",
+      "sparse",
+      "sparse",
+      "sparse"
+    ],
+    "model_type": "glm5_next_text",
+    "moe_intermediate_size": 2048,
+    "moe_router_dtype": "float32",
+    "n_group": 1,
+    "n_routed_experts": 288,
+    "n_shared_experts": 1,
+    "norm_topk_prob": true,
+    "num_attention_heads": 64,
+    "num_experts_per_tok": 8,
+    "num_hidden_layers": 45,
+    "num_key_value_heads": 64,
+    "num_nextn_predict_layers": 1,
+    "output_router_logits": false,
+    "pad_token_id": 154820,
+    "q_lora_rank": 1536,
+    "qk_head_dim": 256,
+    "qk_nope_head_dim": 256,
+    "qk_rope_head_dim": 0,
+    "rms_norm_eps": 1e-05,
+    "routed_scaling_factor": 2.5,
+    "router_aux_loss_coef": 0.001,
+    "scoring_func": "sigmoid",
+    "swiglu_limit": 10.0,
+    "tie_word_embeddings": false,
+    "topk_group": 1,
+    "topk_method": "noaux_tc",
+    "use_cache": true,
+    "v_head_dim": 256,
+    "vocab_size": 154880
+  },
+  "vision_config": {
+    "attention_bias": true,
+    "attention_dropout": 0.0,
+    "depth": 24,
+    "hidden_act": "silu",
+    "hidden_size": 1024,
+    "image_size": 448,
+    "in_channels": 3,
+    "initializer_range": 0.02,
+    "intermediate_size": 4096,
+    "model_type": "glm5_next_vision",
+    "num_heads": 16,
+    "out_hidden_size": 4096,
+    "patch_size": 14,
+    "projection_intermediate_size": 10240,
+    "rms_norm_eps": 1e-05,
+    "spatial_merge_size": 2,
+    "swiglu_limit": 10.0,
+    "temporal_patch_size": 2
+  },
+  "image_token_id": 154854,
+  "video_token_id": 154855,
+  "image_start_token_id": 154830,
+  "image_end_token_id": 154831,
+  "video_start_token_id": 154832,
+  "video_end_token_id": 154833,
+  "tie_word_embeddings": false,
+  "model_type": "glm5_next",
+  "transformers_version": "5.16.0",
+  "quantization": {
+    "group_size": 64,
+    "bits": 4,
+    "language_model.model.layers.3.self_attn.indexer.wq_b": {
+      "group_size": 64,
+      "bits": 8
+    },
+    "language_model.model.layers.3.self_attn.indexer.wk": {
+      "group_size": 64,
+      "bits": 8
+    },
+    "language_model.model.layers.3.self_attn.indexer.weights_proj": {
+      "group_size": 64,
+      "bits": 8
+    },
+    "language_model.model.layers.7.self_attn.indexer.wq_b": {
+      "group_size": 64,
+      "bits": 8
+    },
+    "language_model.model.layers.7.self_attn.indexer.wk": {
+      "group_size": 64,
+      "bits": 8
+    },
+    "language_model.model.layers.7.self_attn.indexer.weights_proj": {
+      "group_size": 64,
+      "bits": 8
+    },
+    "language_model.model.layers.11.self_attn.indexer.wq_b": {
+      "group_size": 64,
+      "bits": 8
+    },
+    "language_model.model.layers.11.self_attn.indexer.wk": {
+      "group_size": 64,
+      "bits": 8
+    },
+    "language_model.model.layers.11.self_attn.indexer.weights_proj": {
+      "group_size": 64,
+      "bits": 8
+    },
+    "language_model.model.layers.15.self_attn.indexer.wq_b": {
+      "group_size": 64,
+      "bits": 8
+    },
+    "language_model.model.layers.15.self_attn.indexer.wk": {
+      "group_size": 64,
+      "bits": 8
+    },
+    "language_model.model.layers.15.self_attn.indexer.weights_proj": {
+      "group_size": 64,
+      "bits": 8
+    },
+    "language_model.model.layers.19.self_attn.indexer.wq_b": {
+      "group_size": 64,
+      "bits": 8
+    },
+    "language_model.model.layers.19.self_attn.indexer.wk": {
+      "group_size": 64,
+      "bits": 8
+    },
+    "language_model.model.layers.19.self_attn.indexer.weights_proj": {
+      "group_size": 64,
+      "bits": 8
+    },
+    "language_model.model.layers.23.self_attn.indexer.wq_b": {
+      "group_size": 64,
+      "bits": 8
+    },
+    "language_model.model.layers.23.self_attn.indexer.wk": {
+      "group_size": 64,
+      "bits": 8
+    },
+    "language_model.model.layers.23.self_attn.indexer.weights_proj": {
+      "group_size": 64,
+      "bits": 8
+    },
+    "language_model.model.layers.27.self_attn.indexer.wq_b": {
+      "group_size": 64,
+      "bits": 8
+    },
+    "language_model.model.layers.27.self_attn.indexer.wk": {
+      "group_size": 64,
+      "bits": 8
+    },
+    "language_model.model.layers.27.self_attn.indexer.weights_proj": {
+      "group_size": 64,
+      "bits": 8
+    },
+    "language_model.model.layers.31.self_attn.indexer.wq_b": {
+      "group_size": 64,
+      "bits": 8
+    },
+    "language_model.model.layers.31.self_attn.indexer.wk": {
+      "group_size": 64,
+      "bits": 8
+    },
+    "language_model.model.layers.31.self_attn.indexer.weights_proj": {
+      "group_size": 64,
+      "bits": 8
+    },
+    "language_model.model.layers.35.self_attn.indexer.wq_b": {
+      "group_size": 64,
+      "bits": 8
+    },
+    "language_model.model.layers.35.self_attn.indexer.wk": {
+      "group_size": 64,
+      "bits": 8
+    },
+    "language_model.model.layers.35.self_attn.indexer.weights_proj": {
+      "group_size": 64,
+      "bits": 8
+    },
+    "language_model.model.layers.39.self_attn.indexer.wq_b": {
+      "group_size": 64,
+      "bits": 8
+    },
+    "language_model.model.layers.39.self_attn.indexer.wk": {
+      "group_size": 64,
+      "bits": 8
+    },
+    "language_model.model.layers.39.self_attn.indexer.weights_proj": {
+      "group_size": 64,
+      "bits": 8
+    },
+    "language_model.model.layers.43.self_attn.indexer.wq_b": {
+      "group_size": 64,
+      "bits": 8
+    },
+    "language_model.model.layers.43.self_attn.indexer.wk": {
+      "group_size": 64,
+      "bits": 8
+    },
+    "language_model.model.layers.43.self_attn.indexer.weights_proj": {
+      "group_size": 64,
+      "bits": 8
+    },
+    "language_model.model.layers.3.mlp.switch_mlp.gate_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.3.mlp.switch_mlp.up_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.3.mlp.switch_mlp.down_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.4.mlp.switch_mlp.gate_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.4.mlp.switch_mlp.up_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.4.mlp.switch_mlp.down_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.5.mlp.switch_mlp.gate_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.5.mlp.switch_mlp.up_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.5.mlp.switch_mlp.down_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.6.mlp.switch_mlp.gate_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.6.mlp.switch_mlp.up_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.6.mlp.switch_mlp.down_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.7.mlp.switch_mlp.gate_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.7.mlp.switch_mlp.up_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.7.mlp.switch_mlp.down_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.8.mlp.switch_mlp.gate_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.8.mlp.switch_mlp.up_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.8.mlp.switch_mlp.down_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.9.mlp.switch_mlp.gate_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.9.mlp.switch_mlp.up_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.9.mlp.switch_mlp.down_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.10.mlp.switch_mlp.gate_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.10.mlp.switch_mlp.up_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.10.mlp.switch_mlp.down_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.11.mlp.switch_mlp.gate_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.11.mlp.switch_mlp.up_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.11.mlp.switch_mlp.down_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.12.mlp.switch_mlp.gate_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.12.mlp.switch_mlp.up_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.12.mlp.switch_mlp.down_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.13.mlp.switch_mlp.gate_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.13.mlp.switch_mlp.up_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.13.mlp.switch_mlp.down_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.14.mlp.switch_mlp.gate_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.14.mlp.switch_mlp.up_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.14.mlp.switch_mlp.down_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.15.mlp.switch_mlp.gate_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.15.mlp.switch_mlp.up_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.15.mlp.switch_mlp.down_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.16.mlp.switch_mlp.gate_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.16.mlp.switch_mlp.up_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.16.mlp.switch_mlp.down_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.17.mlp.switch_mlp.gate_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.17.mlp.switch_mlp.up_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.17.mlp.switch_mlp.down_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.18.mlp.switch_mlp.gate_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.18.mlp.switch_mlp.up_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.18.mlp.switch_mlp.down_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.19.mlp.switch_mlp.gate_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.19.mlp.switch_mlp.up_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.19.mlp.switch_mlp.down_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.20.mlp.switch_mlp.gate_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.20.mlp.switch_mlp.up_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.20.mlp.switch_mlp.down_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.21.mlp.switch_mlp.gate_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.21.mlp.switch_mlp.up_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.21.mlp.switch_mlp.down_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.22.mlp.switch_mlp.gate_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.22.mlp.switch_mlp.up_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.22.mlp.switch_mlp.down_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.23.mlp.switch_mlp.gate_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.23.mlp.switch_mlp.up_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.23.mlp.switch_mlp.down_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.24.mlp.switch_mlp.gate_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.24.mlp.switch_mlp.up_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.24.mlp.switch_mlp.down_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.25.mlp.switch_mlp.gate_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.25.mlp.switch_mlp.up_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.25.mlp.switch_mlp.down_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.26.mlp.switch_mlp.gate_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.26.mlp.switch_mlp.up_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.26.mlp.switch_mlp.down_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.27.mlp.switch_mlp.gate_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.27.mlp.switch_mlp.up_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.27.mlp.switch_mlp.down_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.28.mlp.switch_mlp.gate_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.28.mlp.switch_mlp.up_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.28.mlp.switch_mlp.down_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.29.mlp.switch_mlp.gate_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.29.mlp.switch_mlp.up_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.29.mlp.switch_mlp.down_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.30.mlp.switch_mlp.gate_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.30.mlp.switch_mlp.up_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.30.mlp.switch_mlp.down_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.31.mlp.switch_mlp.gate_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.31.mlp.switch_mlp.up_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.31.mlp.switch_mlp.down_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.32.mlp.switch_mlp.gate_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.32.mlp.switch_mlp.up_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.32.mlp.switch_mlp.down_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.33.mlp.switch_mlp.gate_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.33.mlp.switch_mlp.up_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.33.mlp.switch_mlp.down_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.34.mlp.switch_mlp.gate_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.34.mlp.switch_mlp.up_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.34.mlp.switch_mlp.down_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.35.mlp.switch_mlp.gate_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.35.mlp.switch_mlp.up_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.35.mlp.switch_mlp.down_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.36.mlp.switch_mlp.gate_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.36.mlp.switch_mlp.up_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.36.mlp.switch_mlp.down_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.37.mlp.switch_mlp.gate_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.37.mlp.switch_mlp.up_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.37.mlp.switch_mlp.down_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.38.mlp.switch_mlp.gate_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.38.mlp.switch_mlp.up_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.38.mlp.switch_mlp.down_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.39.mlp.switch_mlp.gate_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.39.mlp.switch_mlp.up_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.39.mlp.switch_mlp.down_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.40.mlp.switch_mlp.gate_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.40.mlp.switch_mlp.up_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.40.mlp.switch_mlp.down_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.41.mlp.switch_mlp.gate_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.41.mlp.switch_mlp.up_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.41.mlp.switch_mlp.down_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.42.mlp.switch_mlp.gate_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.42.mlp.switch_mlp.up_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.42.mlp.switch_mlp.down_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.43.mlp.switch_mlp.gate_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.43.mlp.switch_mlp.up_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.43.mlp.switch_mlp.down_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.44.mlp.switch_mlp.gate_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.44.mlp.switch_mlp.up_proj": {
+      "group_size": 64,
+      "bits": 4
+    },
+    "language_model.model.layers.44.mlp.switch_mlp.down_proj": {
+      "group_size": 64,
+      "bits": 4
+    }
+  },
+  "quantization_config": {
+    "group_size": 64,
+    "bits": 4,
+    "language_model.model.layers.3.self_attn.indexer.wq_b": {
+      "group_size": 64,
+      "bits": 8
+    },
+    "language_model.model.layers.3.self_attn.indexer.wk": {
+      "group_size": 64,
+      "bits": 8
+    },
+    "language_model.model.layers.3.self_attn.indexer.weights_proj": {
+      "group_size": 64,
+      "bits": 8
+    },
+    "language_model.model.layers.7.self_attn.indexer.wq_b": {
+      "group_size": 64,
+      "bits": 8
+    },
+    "language_model.model.layers.7.self_attn.indexer.wk": {
+      "group_size": 64,
+      "bits": 8
+    },
+    "language_model.model.layers.7.self_attn.indexer.weights_proj": {
+      "group_size": 64,
+      "bits": 8
+    },
+    "language_model.model.layers.11.self_attn.indexer.wq_b": {
+      "group_size": 64,
+      "bits": 8
+    },
+    "language_model.model.layers.11.self_attn.indexer.wk": {
+      "group_size": 64,
+      "bits": 8
+    },
+    "language_model.model.layers.11.self_attn.indexer.weights_proj": {
+      "group_size": 64,
+      "bits": 8
+    },
+    "language_model.model.layers.15.self_attn.indexer.wq_b": {
+      "group_size": 64,
+      "bits": 8
+    },
+    "language_model.model.layers.15.self_attn.indexer.wk": {
+      "group_size": 64,
+      "bits": 8
+    },
+    "language_model.model.layers.15.self_attn.indexer.weights_proj": {
+      "group_size": 64,
+      "bits": 8
+    },
+    "language_model.model.layers.19.self_attn.indexer.wq_b": {
+      "group_size": 64,
+      "bits": 8
+    },
+    "language_model.model.layers.19.self_attn.indexer.wk": {
+      "group_size": 64,
+      "bits": 8
+    },
+    "language_model.model.layers.19.self_attn.indexer.weights_proj": {
+      "group_size": 64,
+      "bits": 8
+    },
+    "language_model.model.layers.23.self_attn.indexer.wq_b": {
+      "group_size": 64,
+      "bits": 8
+    },
+    "language_model.model.layers.23.self_attn.indexer.wk": {
+      "group_size": 64,
+      "bits": 8
+    },
+    "language_model.model.layers.23.self_attn.indexer.weights_proj": {
+      "group_size": 64,
+      "bits": 8
+    },
+    "language_model.model.layers.27.self_attn.indexer.wq_b": {
+      "group_size": 64,
+      "bits": 8
+    },
+    "language_model.model.layers.27.self_attn.indexer.wk": {
+      "group_size": 64,
+      "bits": 8
+    },
+    "language_model.model.layers.27.self_attn.indexer.weights_proj": {
+      "group_size": 64,
+      "bits": 8
+    },
+    "language_model.model.layers.31.self_attn.indexer.wq_b": {
+      "group_size": 64,
+      "bits": 8
+    },
+    "language_model.model.layers.31.self_attn.indexer.wk": {
+      "group_size": 64,
+      "bits": 8
+    },
+    "language_model.model.layers.31.self_attn.indexer.weights_proj": {
+      "group_size": 64,
+      "bits": 8
+    },
+    "language_model.model.layers.35.self_attn.indexer.wq_b": {
+      "group_size": 64,
+      "bits": 8
+    },
+    "language_model.model.layers.35.self_attn.indexer.wk": {
+      "group_size": 64,
+      "bits": 8
+    },
+    "language_model.model.layers.35.self_attn.indexer.weights_proj": {
+      "group_size": 64,
+      "bits": 8
+    },
+    "language_model.model.layers.39.self_attn.indexer.wq_b": {
+      "group_size": 64,
+      "bits": 8
+    },
+    "language_model.model.layers.39.self_attn.indexer.wk": {
+      "group_size": 64,
+      "bits": 8
+    },
+    "language_model.model.layers.39.self_attn.indexer.weights_proj": {
+      "group_size": 64,
+      "bits": 8
+    },
+    "language_model.model.layers.43.self_attn.indexer.wq_b": {
+      "group_size": 64,
+      "bits": 8
+    },
+    "language_model.model.layers.43.self_attn.indexer.wk": {
+      "group_size": 64,
+      "bits": 8
+    },
+    "language_model.model.layers.43.self_attn.indexer.weights_proj": {
+      "group_size": 64,
+      "bits": 8
+    }
+  },
+  "gbx_weight_format": "mlx"
+}

experts.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:935c721587cb75373dd88b7554a3a20b1ec79e4e15e7614177c86d50488d8bb5
+size 171228266496

experts_index.json

@@ -0,0 +1,180 @@
+{
+  "format": "gbx-paged-experts-v1",
+  "source": "",
+  "num_layers": 42,
+  "num_experts": 288,
+  "layers": [
+    3,
+    4,
+    5,
+    6,
+    7,
+    8,
+    9,
+    10,
+    11,
+    12,
+    13,
+    14,
+    15,
+    16,
+    17,
+    18,
+    19,
+    20,
+    21,
+    22,
+    23,
+    24,
+    25,
+    26,
+    27,
+    28,
+    29,
+    30,
+    31,
+    32,
+    33,
+    34,
+    35,
+    36,
+    37,
+    38,
+    39,
+    40,
+    41,
+    42,
+    43,
+    44
+  ],
+  "fields": [
+    {
+      "proj": "gate_proj",
+      "field": "weight",
+      "dtype": "uint32",
+      "shape": [
+        2048,
+        512
+      ],
+      "offset": 0,
+      "nbytes": 4194304
+    },
+    {
+      "proj": "gate_proj",
+      "field": "scales",
+      "dtype": "bfloat16",
+      "shape": [
+        2048,
+        64
+      ],
+      "offset": 4194304,
+      "nbytes": 262144
+    },
+    {
+      "proj": "gate_proj",
+      "field": "biases",
+      "dtype": "bfloat16",
+      "shape": [
+        2048,
+        64
+      ],
+      "offset": 4456448,
+      "nbytes": 262144
+    },
+    {
+      "proj": "up_proj",
+      "field": "weight",
+      "dtype": "uint32",
+      "shape": [
+        2048,
+        512
+      ],
+      "offset": 4718592,
+      "nbytes": 4194304
+    },
+    {
+      "proj": "up_proj",
+      "field": "scales",
+      "dtype": "bfloat16",
+      "shape": [
+        2048,
+        64
+      ],
+      "offset": 8912896,
+      "nbytes": 262144
+    },
+    {
+      "proj": "up_proj",
+      "field": "biases",
+      "dtype": "bfloat16",
+      "shape": [
+        2048,
+        64
+      ],
+      "offset": 9175040,
+      "nbytes": 262144
+    },
+    {
+      "proj": "down_proj",
+      "field": "weight",
+      "dtype": "uint32",
+      "shape": [
+        4096,
+        256
+      ],
+      "offset": 9437184,
+      "nbytes": 4194304
+    },
+    {
+      "proj": "down_proj",
+      "field": "scales",
+      "dtype": "bfloat16",
+      "shape": [
+        4096,
+        32
+      ],
+      "offset": 13631488,
+      "nbytes": 262144
+    },
+    {
+      "proj": "down_proj",
+      "field": "biases",
+      "dtype": "bfloat16",
+      "shape": [
+        4096,
+        32
+      ],
+      "offset": 13893632,
+      "nbytes": 262144
+    }
+  ],
+  "payload_bytes": 14155776,
+  "record_bytes": 14155776,
+  "align": 4096,
+  "region": "experts.bin",
+  "created": "2026-09-01T22:41:07",
+  "trunk": "model.safetensors",
+  "self_contained": true,
+  "source_prefix": "language_model.model",
+  "source_id": "pipenetwork/GLM-5.3-Flash-MLX-4bit",
+  "ple_store": null,
+  "verified": {
+    "against": "pipenetwork/GLM-5.3-Flash-MLX-4bit",
+    "at": "2026-09-01T20:45:56+00:00",
+    "slots_per_layer": 103,
+    "gates": [
+      {
+        "gate": "gate 3",
+        "name": "bit-identical logits",
+        "status": "SKIP",
+        "detail": "cannot hold a resident reference: 165.4 GiB checkpoint needs ~198.4 GiB resident, 181.8 GiB usable"
+      },
+      {
+        "gate": "gate 32",
+        "name": "layer-wise vs resident",
+        "status": "PASS",
+        "detail": "42 layers x 2 draws exact, 60.93 GiB peak for this gate"
+      }
+    ]
+  }
+}

generation_config.json

@@ -0,0 +1,12 @@
+{
+  "_from_model_config": true,
+  "eos_token_id": [
+    154820,
+    154827,
+    154829
+  ],
+  "pad_token_id": 154820,
+  "temperature": 1.0,
+  "top_p": 0.95,
+  "transformers_version": "5.16.0"
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e7e06bb445115e322e98695890892f91d1eb0b5fe548076e7521dcdcd9e82405
+size 6320657246

processor_config.json

@@ -0,0 +1,44 @@
+{
+  "image_processor": {
+    "do_rescale": true,
+    "patch_expand_factor": 1,
+    "merge_size": 2,
+    "image_mean": [
+      0.48145466,
+      0.4578275,
+      0.40821073
+    ],
+    "image_std": [
+      0.26862954,
+      0.26130258,
+      0.27577711
+    ],
+    "temporal_patch_size": 2,
+    "patch_size": 14,
+    "min_image_tokens": 16,
+    "max_image_tokens": 8000,
+    "image_processor_type": "Glm5NextImageProcessor"
+  },
+  "video_processor": {
+    "do_rescale": true,
+    "video_processor_type": "Glm5NextVideoProcessor",
+    "patch_expand_factor": 1,
+    "merge_size": 2,
+    "image_mean": [
+      0.48145466,
+      0.4578275,
+      0.40821073
+    ],
+    "image_std": [
+      0.26862954,
+      0.26130258,
+      0.27577711
+    ],
+    "temporal_patch_size": 2,
+    "patch_size": 14,
+    "min_image_tokens": 16,
+    "max_image_tokens": 240000,
+    "fps": 2
+  },
+  "processor_class": "Glm5NextProcessor"
+}

tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:19e773648cb4e65de8660ea6365e10acca112d42a854923df93db4a6f333a82d
+size 20217442

tokenizer_config.json

@@ -0,0 +1,33 @@
+{
+  "backend": "tokenizers",
+  "clean_up_tokenization_spaces": false,
+  "do_lower_case": false,
+  "eos_token": "<|endoftext|>",
+  "extra_special_tokens": [
+    "<|endoftext|>",
+    "[MASK]",
+    "[gMASK]",
+    "[sMASK]",
+    "<sop>",
+    "<eop>",
+    "<|system|>",
+    "<|user|>",
+    "<|assistant|>",
+    "<|observation|>",
+    "<|begin_of_image|>",
+    "<|end_of_image|>",
+    "<|begin_of_video|>",
+    "<|end_of_video|>",
+    "<|begin_of_audio|>",
+    "<|end_of_audio|>",
+    "<|begin_of_transcription|>",
+    "<|end_of_transcription|>"
+  ],
+  "is_local": true,
+  "model_max_length": 1048576,
+  "model_specific_special_tokens": {},
+  "pad_token": "<|endoftext|>",
+  "padding_side": "left",
+  "remove_space": false,
+  "tokenizer_class": "TokenizersBackend"
+}