Update bundle from mazesmazes/tiny-audio-embedded

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* 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
+tokenizer.json filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,19 @@
+# tiny-audio Swift bundle
+
+MLX bundle for [tiny-audio-swift](https://github.com/alexkroman/tiny-audio-swift), built from projector [`mazesmazes/tiny-audio-embedded`](https://huggingface.co/mazesmazes/tiny-audio-embedded).
+
+## Contents
+- `encoder.safetensors` — GLM-ASR encoder, quantized 8-bit (group 64)
+- `projector.safetensors` — fp16 projector
+- `decoder.safetensors` — MLX-LM decoder
+- `config.json`, `tokenizer.json`, `tokenizer_config.json`, `decoder_config.json`
+- `manifest.json` — sha256 + sizes for all files
+
+## Model config
+- Encoder dim: 1280, layers: 32, heads: 20
+- Audio token: `<audio>` (id 151669)
+- Hop length: 160
+- Format version: 1
+
+## Usage
+Loaded automatically by tiny-audio-swift's `Transcriber`.

config.json

@@ -0,0 +1,41 @@
+{
+  "mlx_format_version": 1,
+  "encoder": {
+    "encoder_dim": 1280,
+    "n_mels": 128,
+    "num_layers": 32,
+    "num_heads": 20,
+    "head_dim": 64,
+    "intermediate_size": 5120,
+    "rope_theta": 10000.0,
+    "quantization": {
+      "group_size": 64,
+      "bits": 8
+    }
+  },
+  "projector": {
+    "encoder_dim": 1280,
+    "llm_dim": 1024,
+    "hidden_dim": 2048,
+    "pool_stride": 4
+  },
+  "audio_token": "<audio>",
+  "audio_token_id": 151669,
+  "eos_token_ids": [
+    151643,
+    151645
+  ],
+  "encoder_conv_layers": [
+    [
+      1,
+      3,
+      1
+    ],
+    [
+      1,
+      3,
+      2
+    ]
+  ],
+  "hop_length": 160
+}

decoder.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e0af2959045dad826ab3dda66dbece3440d049f6115d6022336b0b52b9afbac8
+size 670516209

decoder_config.json

@@ -0,0 +1,77 @@
+{
+    "architectures": [
+        "Qwen3ForCausalLM"
+    ],
+    "attention_bias": false,
+    "attention_dropout": 0.0,
+    "bos_token_id": null,
+    "dtype": "float32",
+    "eos_token_id": [
+        151645,
+        151643
+    ],
+    "head_dim": 128,
+    "hidden_act": "silu",
+    "hidden_size": 1024,
+    "initializer_range": 0.02,
+    "intermediate_size": 3072,
+    "layer_types": [
+        "full_attention",
+        "full_attention",
+        "full_attention",
+        "full_attention",
+        "full_attention",
+        "full_attention",
+        "full_attention",
+        "full_attention",
+        "full_attention",
+        "full_attention",
+        "full_attention",
+        "full_attention",
+        "full_attention",
+        "full_attention",
+        "full_attention",
+        "full_attention",
+        "full_attention",
+        "full_attention",
+        "full_attention",
+        "full_attention",
+        "full_attention",
+        "full_attention",
+        "full_attention",
+        "full_attention",
+        "full_attention",
+        "full_attention",
+        "full_attention",
+        "full_attention"
+    ],
+    "max_position_embeddings": 40960,
+    "max_window_layers": 28,
+    "model_type": "qwen3",
+    "num_attention_heads": 16,
+    "num_hidden_layers": 28,
+    "num_key_value_heads": 8,
+    "pad_token_id": 151643,
+    "quantization": {
+        "group_size": 64,
+        "bits": 8,
+        "mode": "affine"
+    },
+    "quantization_config": {
+        "group_size": 64,
+        "bits": 8,
+        "mode": "affine"
+    },
+    "rms_norm_eps": 1e-06,
+    "rope_parameters": {
+        "rope_theta": 1000000,
+        "rope_type": "default"
+    },
+    "rope_theta": 1000000.0,
+    "sliding_window": null,
+    "tie_word_embeddings": true,
+    "transformers_version": "5.7.0",
+    "use_cache": false,
+    "use_sliding_window": false,
+    "vocab_size": 151670
+}

encoder.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:40ae3b9cdc513670685eec797168a80a78dd9808ea5a10e25177e03213cf063e
+size 682529665

manifest.json

@@ -0,0 +1,37 @@
+{
+  "format_version": 1,
+  "files": {
+    "tokenizer_config.json": {
+      "sha256": "2b682cdd791448785caf44a1da2bafffdecaaf88ae6654fd8c3b59cc3b7470ab",
+      "size": 4766
+    },
+    "decoder_config.json": {
+      "sha256": "a8374eef7dacf1580738fd8ddc1f7e0b9e948866e1dd0d044f17eb85a3a42014",
+      "size": 1865
+    },
+    "config.json": {
+      "sha256": "c63e8bf5ea9b76fbcd734694d62e2b75ba6d6483fd2881cd18b00b71b2ab76bf",
+      "size": 630
+    },
+    "tokenizer.json": {
+      "sha256": "33b674fb8444e2553eae8f1b261093371920a28ef75b5c18f4deb3f9217ed0ba",
+      "size": 11422834
+    },
+    "README.md": {
+      "sha256": "c38020c601a9d28abb1dc01661161ffb0a3c8cf362ce2bec029aeaa53437115b",
+      "size": 928
+    },
+    "projector.safetensors": {
+      "sha256": "44e021cee2070bffabf612ff1f02028f58948dc854ad69b743bc8847314a9f04",
+      "size": 50344282
+    },
+    "encoder.safetensors": {
+      "sha256": "40ae3b9cdc513670685eec797168a80a78dd9808ea5a10e25177e03213cf063e",
+      "size": 682529665
+    },
+    "decoder.safetensors": {
+      "sha256": "e0af2959045dad826ab3dda66dbece3440d049f6115d6022336b0b52b9afbac8",
+      "size": 670516209
+    }
+  }
+}

projector.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:44e021cee2070bffabf612ff1f02028f58948dc854ad69b743bc8847314a9f04
+size 50344282

tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:33b674fb8444e2553eae8f1b261093371920a28ef75b5c18f4deb3f9217ed0ba
+size 11422834

tokenizer_config.json

@@ -0,0 +1,20 @@
+{
+  "add_prefix_space": false,
+  "backend": "tokenizers",
+  "bos_token": null,
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "errors": "replace",
+  "extra_special_tokens": [
+    "<audio>"
+  ],
+  "is_local": true,
+  "local_files_only": false,
+  "model_max_length": 131072,
+  "pad_token": "<|endoftext|>",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "tool_parser_type": "json_tools",
+  "unk_token": null,
+  "chat_template": "{%- if tools %}\n    {{- '<|im_start|>system\\n' }}\n    {%- if messages[0].role == 'system' %}\n        {{- messages[0].content + '\\n\\n' }}\n    {%- endif %}\n    {{- \"# Tools\\n\\nYou may call one or more functions to assist with the user query.\\n\\nYou are provided with function signatures within <tools></tools> XML tags:\\n<tools>\" }}\n    {%- for tool in tools %}\n        {{- \"\\n\" }}\n        {{- tool | tojson }}\n    {%- endfor %}\n    {{- \"\\n</tools>\\n\\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\\n<tool_call>\\n{\\\"name\\\": <function-name>, \\\"arguments\\\": <args-json-object>}\\n</tool_call><|im_end|>\\n\" }}\n{%- else %}\n    {%- if messages[0].role == 'system' %}\n        {{- '<|im_start|>system\\n' + messages[0].content + '<|im_end|>\\n' }}\n    {%- endif %}\n{%- endif %}\n{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}\n{%- for message in messages[::-1] %}\n    {%- set index = (messages|length - 1) - loop.index0 %}\n    {%- if ns.multi_step_tool and message.role == \"user\" and message.content is string and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}\n        {%- set ns.multi_step_tool = false %}\n        {%- set ns.last_query_index = index %}\n    {%- endif %}\n{%- endfor %}\n{%- for message in messages %}\n    {%- if message.content is string %}\n        {%- set content = message.content %}\n    {%- else %}\n        {%- set content = '' %}\n    {%- endif %}\n    {%- if (message.role == \"user\") or (message.role == \"system\" and not loop.first) %}\n        {{- '<|im_start|>' + message.role + '\\n' + content + '<|im_end|>' + '\\n' }}\n    {%- elif message.role == \"assistant\" %}\n        {%- set reasoning_content = '' %}\n        {%- if message.reasoning_content is string %}\n            {%- set reasoning_content = message.reasoning_content %}\n        {%- else %}\n            {%- if '</think>' in content %}\n                {%- set reasoning_content = content.split('</think>')[0].rstrip('\\n').split('<think>')[-1].lstrip('\\n') %}\n                {%- set content = content.split('</think>')[-1].lstrip('\\n') %}\n            {%- endif %}\n        {%- endif %}\n        {%- if loop.index0 > ns.last_query_index %}\n            {%- if loop.last or (not loop.last and reasoning_content) %}\n                {{- '<|im_start|>' + message.role + '\\n<think>\\n' + reasoning_content.strip('\\n') + '\\n</think>\\n\\n' + content.lstrip('\\n') }}\n            {%- else %}\n                {{- '<|im_start|>' + message.role + '\\n' + content }}\n            {%- endif %}\n        {%- else %}\n            {{- '<|im_start|>' + message.role + '\\n' + content }}\n        {%- endif %}\n        {%- if message.tool_calls %}\n            {%- for tool_call in message.tool_calls %}\n                {%- if (loop.first and content) or (not loop.first) %}\n                    {{- '\\n' }}\n                {%- endif %}\n                {%- if tool_call.function %}\n                    {%- set tool_call = tool_call.function %}\n                {%- endif %}\n                {{- '<tool_call>\\n{\"name\": \"' }}\n                {{- tool_call.name }}\n                {{- '\", \"arguments\": ' }}\n                {%- if tool_call.arguments is string %}\n                    {{- tool_call.arguments }}\n                {%- else %}\n                    {{- tool_call.arguments | tojson }}\n                {%- endif %}\n                {{- '}\\n</tool_call>' }}\n            {%- endfor %}\n        {%- endif %}\n        {{- '<|im_end|>\\n' }}\n    {%- elif message.role == \"tool\" %}\n        {%- if loop.first or (messages[loop.index0 - 1].role != \"tool\") %}\n            {{- '<|im_start|>user' }}\n        {%- endif %}\n        {{- '\\n<tool_response>\\n' }}\n        {{- content }}\n        {{- '\\n</tool_response>' }}\n        {%- if loop.last or (messages[loop.index0 + 1].role != \"tool\") %}\n            {{- '<|im_end|>\\n' }}\n        {%- endif %}\n    {%- endif %}\n{%- endfor %}\n{%- if add_generation_prompt %}\n    {{- '<|im_start|>assistant\\n' }}\n    {%- if true %}\n        {{- '<think>\\n\\n</think>\\n\\n' }}\n    {%- endif %}\n{%- endif %}"
+}