Dots.tts (code, models, paper)

.gitattributes

@@ -33,3 +33,11 @@ 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
+Dots.tts[[:space:]]Technical[[:space:]]Report.pdf filter=lfs diff=lfs merge=lfs -text
+models/dots-tts-mlx/int4/tokenizer/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+models/dots-tts-mlx/int8/tokenizer/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+models/dots-tts-mlx/mf-int4/tokenizer/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+models/dots-tts-mlx/mf-int8/tokenizer/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+models/dots.tts-base/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+models/dots.tts-mf/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+models/dots.tts-soar/tokenizer.json filter=lfs diff=lfs merge=lfs -text

Dots.tts Technical Report.pdf

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3ab35d1e2b10269573897fe2d98fa5471c84e24647c4739c4b3862b45b15b357
+size 321356

code/Dots-TTS-ComfyUI.zip

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c9511554ab48c2088b3e845f6782693d812b5c82e8bb5289f1ee8bb4bb743ce2
+size 16788031

code/dots-tts-launcher.zip

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f4a3e7bc35b8b8fe67d620b825d64e626990c77325d1db5ba4fd63dc5f9aacc7
+size 49600

code/dots-tts-mlx.zip

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6b5b050d4085d6f3dd884fece9713ca1ede8bb2893b76ad2598e45812a612a73
+size 436577

code/dots-tts-runpod.zip

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c5389cc178a4e60a72e0b6ba9d340bd92c7bec8dce601c499cdc1608d97f5261
+size 65924

code/dots.tts.cpp.zip

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a5dd676b263fb3c142e699fae3788d03fdea37cbfdffd3cb8729b44aa3ab9d99
+size 76041325

code/dots.tts.zip

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f9fc70b31623ebfcd6ee4d16c5770308a882c3a93b3aa1a3aebff9e499070075
+size 1178074

models/dots-tts-mlx/.gitattributes

@@ -0,0 +1,39 @@
+*.7z filter=lfs diff=lfs merge=lfs -text
+*.arrow 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
+int4/tokenizer/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+int8/tokenizer/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+mf-int4/tokenizer/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+mf-int8/tokenizer/tokenizer.json filter=lfs diff=lfs merge=lfs -text

models/dots-tts-mlx/README.md

@@ -0,0 +1,104 @@
+---
+license: apache-2.0
+base_model:
+  - rednote-hilab/dots.tts-soar
+  - rednote-hilab/dots.tts-mf
+pipeline_tag: text-to-speech
+library_name: mlx
+tags:
+  - mlx
+  - text-to-speech
+  - tts
+  - voice-cloning
+  - apple-silicon
+  - dots.tts
+language:
+  - ar
+  - yue
+  - zh
+  - cs
+  - nl
+  - en
+  - fi
+  - fr
+  - de
+  - el
+  - hi
+  - id
+  - it
+  - ja
+  - ko
+  - pl
+  - pt
+  - ro
+  - ru
+  - es
+  - th
+  - tr
+  - uk
+  - vi
+---
+
+# dots-tts-mlx — quantized MLX weights (Apple Silicon)
+
+Ready-to-run **MLX** weights for [`rednote-hilab/dots.tts-soar`](https://huggingface.co/rednote-hilab/dots.tts-soar) — a 2B continuous-AR flow-matching, **multilingual (24 languages, same as upstream)**, zero-shot voice-clone TTS — quantized for Apple Silicon. **Download and run with the [`dots-tts-mlx`](https://github.com/sb1992/dots-tts-mlx) runtime — no PyTorch and no conversion step.**
+
+> **Languages:** same as upstream dots.tts — **all 24** (Arabic, Cantonese, Chinese, Czech, Dutch, English, Finnish, French, German, Greek, Hindi, Indonesian, Italian, Japanese, Korean, Polish, Portuguese, Romanian, Russian, Spanish, Thai, Turkish, Ukrainian, Vietnamese). The 5-language check in [Quality](#quality) is only the **quantization spot-check**, not the supported set.
+
+> These are converted + LLM-quantized **MLX** safetensors, not PyTorch. They load only with the `dots-tts-mlx` runtime on Apple Silicon (Metal). For the original PyTorch model, see [rednote-hilab/dots.tts-soar](https://huggingface.co/rednote-hilab/dots.tts-soar).
+
+> ⚡ **Two decoders, one voice.** `int4`/`int8` are the standard **soar** build (reference quality). `mf-int4`/`mf-int8` are **MeanFlow** — a distilled few-step decoder (`rednote-hilab/dots.tts-mf`) that runs the acoustic DiT at **NFE=4 with no classifier-free guidance**, **~2× faster** than the 10-step path with no measurable quality loss. Same model, same 24-language cloning — pick the folder that fits your speed budget. For cloning, **reference mode (audio + transcript) is the recommended quality path.**
+
+## Variants
+
+| Subfolder | Decoder | Download | Speed | Use |
+|-----------|---------|----------|-------|-----|
+| `int4/` ⭐ | soar — 10-step + CFG | ~2.4 GB | baseline | default, best quality |
+| `int8/` | soar — 10-step + CFG | ~3.1 GB | baseline | conservative fallback |
+| `mf-int4/` ⚡ | **MeanFlow — NFE-4, no CFG** | ~2.4 GB | **~2× faster** | latency-sensitive |
+| `mf-int8/` | **MeanFlow — NFE-4, no CFG** | ~3.1 GB | **~2× faster** | meanflow + more LLM precision |
+
+Only the **Qwen2.5-1.5B LLM trunk** (≈70% of the weights) is quantized (group-wise affine, group size 64); the precision-sensitive flow-matching DiT, the BigVGAN vocoder, and the CAM++ speaker encoder stay **bf16**.
+
+**MeanFlow** (`mf-*`) is the distilled [`rednote-hilab/dots.tts-mf`](https://huggingface.co/rednote-hilab/dots.tts-mf) checkpoint — the same architecture plus a small duration embedder — **auto-detected** from `config.json` (no flag): point the runtime at an `mf-*` folder and it uses the few-step solver. Measured **1.9–2.2× faster** than the 10-step path on reference cloning (EN/HI/ZH). It drops classifier-free guidance, so `--guidance-scale` is ignored. Use **reference cloning (audio + transcript)** for best quality.
+
+## Quality
+
+Quantization is validated to be **lossless relative to the full-precision MLX build**: on a small multilingual acceptance check (EN/DE/ES/FR + Hindi), int8 and int4 showed **no transcription-accuracy or voice-similarity regression vs bf16**. This is a sanity check, **not a dataset-scale benchmark** — evaluate on your own content.
+
+Correctness of the port itself is gated **per-stage** against the original PyTorch model (AudioVAE PSNR ≈ 56 dB; attention / DiT / LLM / semantic-encoder cosine ≥ 0.9999) — see the [runtime repo](https://github.com/sb1992/dots-tts-mlx).
+
+## Usage
+
+```bash
+# 1. install the quant-aware runtime (>= v0.2.0)
+pip install "git+https://github.com/sb1992/dots-tts-mlx.git@v0.2.0"
+
+# 2. download the variant you want  (use "mf-int4/*" for the faster MeanFlow decoder)
+hf download shraey/dots-tts-mlx --include "int4/*" --local-dir ./dots-tts-mlx-weights
+
+# 3. run (files land in ./dots-tts-mlx-weights/int4/)
+dots-tts --model ./dots-tts-mlx-weights/int4 \
+    --text "Hello from MLX." --ref-audio reference.wav --language EN \
+    --out-path out --out-prefix clone
+```
+
+The runtime auto-detects the `quantization` block in `config.json`, so nothing changes at the CLI/API level vs an unquantized directory. Python API and the full flag set: see the [runtime repo](https://github.com/sb1992/dots-tts-mlx).
+
+- **Memory:** peak RAM scales with generation + reference length — roughly **~6 GB for a short clip, up to ~13 GB for a ~30 s clip** (int4); resident weights are ~2.4 GB. The render peak is activation-bound (the bf16 DiT + vocoder working set), so it's the same for soar and MeanFlow and isn't reduced by quantization. MLX's allocator may cache up to its memory limit, but that cache is releasable.
+- **Requires:** Apple Silicon (MLX is Metal-only), Python ≥ 3.10.
+
+## Attribution & licenses
+
+Derivative quantized weights of `rednote-hilab/dots.tts-soar` (**Apache-2.0**) — you must comply with the upstream license. Components:
+
+- **dots.tts** — [model](https://huggingface.co/rednote-hilab/dots.tts-soar) · [code](https://github.com/rednote-hilab/dots.tts) — Apache-2.0, © the dots.tts team at rednote-hilab.
+- **Qwen2.5-1.5B-Base** (LLM backbone) — Apache-2.0.
+- **CAM++ / 3D-Speaker** (speaker x-vector encoder) — Apache-2.0.
+- **BigVGAN** (vocoder/decoder architecture style) — MIT, © NVIDIA.
+
+MLX port + quantization code: [github.com/sb1992/dots-tts-mlx](https://github.com/sb1992/dots-tts-mlx) (Apache-2.0).
+
+## Responsible use
+
+This performs **zero-shot voice cloning** — it can reproduce a person's voice from a few seconds of audio. Only clone voices you own or for which you have explicit, informed consent; do not use it for impersonation, fraud, or deception; and disclose AI-generated audio wherever it's shared. See the upstream [risks guidance](https://huggingface.co/rednote-hilab/dots.tts-soar).

models/dots-tts-mlx/int4/config.json

@@ -0,0 +1,120 @@
+{
+  "model_type": "dots_tts",
+  "latent_dim": 128,
+  "patch_size": 4,
+  "cfg_droprate": 0.2,
+  "PatchEncoder": {
+    "num_layers": 24,
+    "num_heads": 16,
+    "hidden_size": 1024,
+    "ffn_hidden_size": 4096,
+    "modulation": false,
+    "qkv_bias": false,
+    "qk_norm": true,
+    "attn_dropout": 0.0,
+    "dropout": 0.1,
+    "norm_layer": "RMSNorm",
+    "alibi_bias": false,
+    "rotary_bias": true,
+    "rotary_theta": 10000.0,
+    "input_dim": 128,
+    "causal": true
+  },
+  "DiT": {
+    "num_layers": 18,
+    "num_heads": 16,
+    "hidden_size": 1024,
+    "ffn_hidden_size": 4096,
+    "modulation": true,
+    "qkv_bias": false,
+    "qk_norm": true,
+    "attn_dropout": 0.0,
+    "dropout": 0.0,
+    "norm_layer": "RMSNorm",
+    "alibi_bias": false,
+    "rotary_bias": true,
+    "rotary_theta": 10000.0
+  },
+  "vocoder": {
+    "sample_rate": 48000,
+    "upsample_rates": [
+      10,
+      6,
+      4,
+      2,
+      2,
+      2
+    ],
+    "upsample_kernel_sizes": [
+      20,
+      12,
+      8,
+      4,
+      4,
+      4
+    ],
+    "upsample_initial_channel": 1536,
+    "resblock": "1",
+    "resblock_kernel_sizes": [
+      3,
+      7,
+      11
+    ],
+    "resblock_dilation_sizes": [
+      [
+        1,
+        3,
+        5
+      ],
+      [
+        1,
+        3,
+        5
+      ],
+      [
+        1,
+        3,
+        5
+      ]
+    ],
+    "downsample_rates": [
+      2,
+      2,
+      2,
+      4,
+      6,
+      10
+    ],
+    "downsample_channels": [
+      12,
+      24,
+      48,
+      96,
+      192,
+      384,
+      768
+    ],
+    "activation": "snakebeta",
+    "snake_logscale": true,
+    "latent_dim": 128,
+    "causal": true,
+    "mi_num_layers": 4,
+    "causal_encoder": true,
+    "use_bias_at_final": false,
+    "use_tanh_at_final": false
+  },
+  "fm_sigma": 0.0,
+  "xvec_drop_rate": 0.2,
+  "campplus_embedding_size": 512,
+  "xvec_max_audio_seconds": 10.0,
+  "architectures": [
+    "DotsTTSForConditionalGeneration"
+  ],
+  "quantization": {
+    "bits": 4,
+    "group_size": 64,
+    "components": [
+      "llm"
+    ]
+  }
+}

models/dots-tts-mlx/int4/core.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6cc6b774262487136911574b138da085e26aaa6895673780d29d5b0ef66396b4
+size 2178035679

models/dots-tts-mlx/int4/latent_stats.npz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2fb11a77b710d3230345bd9b0985c075168832ef4b819af317159f31673aa21d
+size 1524

models/dots-tts-mlx/int4/llm_config.json

@@ -0,0 +1,114 @@
+{
+  "vocab_size": 151672,
+  "max_position_embeddings": 131072,
+  "hidden_size": 1536,
+  "intermediate_size": 8960,
+  "num_hidden_layers": 28,
+  "num_attention_heads": 12,
+  "use_sliding_window": false,
+  "sliding_window": null,
+  "max_window_layers": 28,
+  "num_key_value_heads": 2,
+  "hidden_act": "silu",
+  "initializer_range": 0.02,
+  "rms_norm_eps": 1e-06,
+  "use_cache": true,
+  "rope_theta": 1000000.0,
+  "rope_scaling": null,
+  "attention_dropout": 0.0,
+  "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"
+  ],
+  "return_dict": true,
+  "output_hidden_states": false,
+  "torchscript": false,
+  "dtype": "bfloat16",
+  "pruned_heads": {},
+  "tie_word_embeddings": true,
+  "chunk_size_feed_forward": 0,
+  "is_encoder_decoder": false,
+  "is_decoder": false,
+  "cross_attention_hidden_size": null,
+  "add_cross_attention": false,
+  "tie_encoder_decoder": false,
+  "architectures": [
+    "Qwen2ForCausalLM"
+  ],
+  "finetuning_task": null,
+  "id2label": {
+    "0": "LABEL_0",
+    "1": "LABEL_1"
+  },
+  "label2id": {
+    "LABEL_0": 0,
+    "LABEL_1": 1
+  },
+  "task_specific_params": null,
+  "problem_type": null,
+  "tokenizer_class": null,
+  "prefix": null,
+  "bos_token_id": 151643,
+  "pad_token_id": null,
+  "eos_token_id": 151643,
+  "sep_token_id": null,
+  "decoder_start_token_id": null,
+  "max_length": 20,
+  "min_length": 0,
+  "do_sample": false,
+  "early_stopping": false,
+  "num_beams": 1,
+  "temperature": 1.0,
+  "top_k": 50,
+  "top_p": 1.0,
+  "typical_p": 1.0,
+  "repetition_penalty": 1.0,
+  "length_penalty": 1.0,
+  "no_repeat_ngram_size": 0,
+  "encoder_no_repeat_ngram_size": 0,
+  "bad_words_ids": null,
+  "num_return_sequences": 1,
+  "output_scores": false,
+  "return_dict_in_generate": false,
+  "forced_bos_token_id": null,
+  "forced_eos_token_id": null,
+  "remove_invalid_values": false,
+  "exponential_decay_length_penalty": null,
+  "suppress_tokens": null,
+  "begin_suppress_tokens": null,
+  "num_beam_groups": 1,
+  "diversity_penalty": 0.0,
+  "_name_or_path": "",
+  "transformers_version": "4.57.0",
+  "model_type": "qwen2",
+  "use_mrope": false,
+  "tf_legacy_loss": false,
+  "use_bfloat16": false,
+  "output_attentions": false
+}

models/dots-tts-mlx/int4/speaker.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bb418d38516cef3266c7c3cdf3da545f5c32ef1c477d629cc0c8b6a214d855c8
+size 14633169

models/dots-tts-mlx/int4/tokenizer/added_tokens.json

@@ -0,0 +1,31 @@
+{
+  "</tool_call>": 151658,
+  "<tool_call>": 151657,
+  "<|audio_comp_end|>": 151667,
+  "<|audio_comp_span|>": 151666,
+  "<|audio_comp_start|>": 151665,
+  "<|audio_gen_end|>": 151670,
+  "<|audio_gen_span|>": 151669,
+  "<|audio_gen_start|>": 151668,
+  "<|box_end|>": 151649,
+  "<|box_start|>": 151648,
+  "<|endoftext|>": 151643,
+  "<|file_sep|>": 151664,
+  "<|fim_middle|>": 151660,
+  "<|fim_pad|>": 151662,
+  "<|fim_prefix|>": 151659,
+  "<|fim_suffix|>": 151661,
+  "<|im_end|>": 151645,
+  "<|im_start|>": 151644,
+  "<|image_pad|>": 151655,
+  "<|object_ref_end|>": 151647,
+  "<|object_ref_start|>": 151646,
+  "<|quad_end|>": 151651,
+  "<|quad_start|>": 151650,
+  "<|repo_name|>": 151663,
+  "<|text_cond_end|>": 151671,
+  "<|video_pad|>": 151656,
+  "<|vision_end|>": 151653,
+  "<|vision_pad|>": 151654,
+  "<|vision_start|>": 151652
+}

models/dots-tts-mlx/int4/tokenizer/chat_template.jinja

@@ -0,0 +1,54 @@
+{%- if tools %}
+    {{- '<|im_start|>system\n' }}
+    {%- if messages[0]['role'] == 'system' %}
+        {{- messages[0]['content'] }}
+    {%- else %}
+        {{- 'You are a helpful assistant.' }}
+    {%- endif %}
+    {{- "\n\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>" }}
+    {%- for tool in tools %}
+        {{- "\n" }}
+        {{- tool | tojson }}
+    {%- endfor %}
+    {{- "\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" }}
+{%- else %}
+    {%- if messages[0]['role'] == 'system' %}
+        {{- '<|im_start|>system\n' + messages[0]['content'] + '<|im_end|>\n' }}
+    {%- else %}
+        {{- '<|im_start|>system\nYou are a helpful assistant.<|im_end|>\n' }}
+    {%- endif %}
+{%- endif %}
+{%- for message in messages %}
+    {%- if (message.role == "user") or (message.role == "system" and not loop.first) or (message.role == "assistant" and not message.tool_calls) %}
+        {{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>' + '\n' }}
+    {%- elif message.role == "assistant" %}
+        {{- '<|im_start|>' + message.role }}
+        {%- if message.content %}
+            {{- '\n' + message.content }}
+        {%- endif %}
+        {%- for tool_call in message.tool_calls %}
+            {%- if tool_call.function is defined %}
+                {%- set tool_call = tool_call.function %}
+            {%- endif %}
+            {{- '\n<tool_call>\n{"name": "' }}
+            {{- tool_call.name }}
+            {{- '", "arguments": ' }}
+            {{- tool_call.arguments | tojson }}
+            {{- '}\n</tool_call>' }}
+        {%- endfor %}
+        {{- '<|im_end|>\n' }}
+    {%- elif message.role == "tool" %}
+        {%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != "tool") %}
+            {{- '<|im_start|>user' }}
+        {%- endif %}
+        {{- '\n<tool_response>\n' }}
+        {{- message.content }}
+        {{- '\n</tool_response>' }}
+        {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
+            {{- '<|im_end|>\n' }}
+        {%- endif %}
+    {%- endif %}
+{%- endfor %}
+{%- if add_generation_prompt %}
+    {{- '<|im_start|>assistant\n' }}
+{%- endif %}

models/dots-tts-mlx/int4/tokenizer/special_tokens_map.json

@@ -0,0 +1,67 @@
+{
+  "additional_special_tokens": [
+    {
+      "content": "<|audio_comp_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false
+    },
+    {
+      "content": "<|audio_comp_span|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false
+    },
+    {
+      "content": "<|audio_comp_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false
+    },
+    {
+      "content": "<|audio_gen_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false
+    },
+    {
+      "content": "<|audio_gen_span|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false
+    },
+    {
+      "content": "<|audio_gen_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false
+    },
+    {
+      "content": "<|text_cond_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false
+    }
+  ],
+  "eos_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

models/dots-tts-mlx/int4/tokenizer/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c16521f66774c7a4774e5303b7c8ec5c99830c0be5aef6c6edde3ca2a5e05dd0
+size 11423263

models/dots-tts-mlx/int4/tokenizer/tokenizer_config.json

@@ -0,0 +1,257 @@
+{
+  "add_bos_token": false,
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "151643": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151644": {
+      "content": "<|im_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151645": {
+      "content": "<|im_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151646": {
+      "content": "<|object_ref_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151647": {
+      "content": "<|object_ref_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151648": {
+      "content": "<|box_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151649": {
+      "content": "<|box_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151650": {
+      "content": "<|quad_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151651": {
+      "content": "<|quad_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151652": {
+      "content": "<|vision_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151653": {
+      "content": "<|vision_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151654": {
+      "content": "<|vision_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151655": {
+      "content": "<|image_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151656": {
+      "content": "<|video_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151657": {
+      "content": "<tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151658": {
+      "content": "</tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151659": {
+      "content": "<|fim_prefix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151660": {
+      "content": "<|fim_middle|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151661": {
+      "content": "<|fim_suffix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151662": {
+      "content": "<|fim_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151663": {
+      "content": "<|repo_name|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151664": {
+      "content": "<|file_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151665": {
+      "content": "<|audio_comp_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151666": {
+      "content": "<|audio_comp_span|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151667": {
+      "content": "<|audio_comp_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151668": {
+      "content": "<|audio_gen_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151669": {
+      "content": "<|audio_gen_span|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151670": {
+      "content": "<|audio_gen_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151671": {
+      "content": "<|text_cond_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "additional_special_tokens": [
+    "<|audio_comp_start|>",
+    "<|audio_comp_span|>",
+    "<|audio_comp_end|>",
+    "<|audio_gen_start|>",
+    "<|audio_gen_span|>",
+    "<|audio_gen_end|>",
+    "<|text_cond_end|>"
+  ],
+  "bos_token": null,
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|endoftext|>",
+  "errors": "replace",
+  "extra_special_tokens": {},
+  "model_max_length": 131072,
+  "pad_token": "<|endoftext|>",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

models/dots-tts-mlx/int4/vocoder.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9786f810cc674ea13f84c0a163186d6d73ce2c0edbb63b08aaacccd90cb613a1
+size 361798651

models/dots-tts-mlx/int8/config.json

@@ -0,0 +1,120 @@
+{
+  "model_type": "dots_tts",
+  "latent_dim": 128,
+  "patch_size": 4,
+  "cfg_droprate": 0.2,
+  "PatchEncoder": {
+    "num_layers": 24,
+    "num_heads": 16,
+    "hidden_size": 1024,
+    "ffn_hidden_size": 4096,
+    "modulation": false,
+    "qkv_bias": false,
+    "qk_norm": true,
+    "attn_dropout": 0.0,
+    "dropout": 0.1,
+    "norm_layer": "RMSNorm",
+    "alibi_bias": false,
+    "rotary_bias": true,
+    "rotary_theta": 10000.0,
+    "input_dim": 128,
+    "causal": true
+  },
+  "DiT": {
+    "num_layers": 18,
+    "num_heads": 16,
+    "hidden_size": 1024,
+    "ffn_hidden_size": 4096,
+    "modulation": true,
+    "qkv_bias": false,
+    "qk_norm": true,
+    "attn_dropout": 0.0,
+    "dropout": 0.0,
+    "norm_layer": "RMSNorm",
+    "alibi_bias": false,
+    "rotary_bias": true,
+    "rotary_theta": 10000.0
+  },
+  "vocoder": {
+    "sample_rate": 48000,
+    "upsample_rates": [
+      10,
+      6,
+      4,
+      2,
+      2,
+      2
+    ],
+    "upsample_kernel_sizes": [
+      20,
+      12,
+      8,
+      4,
+      4,
+      4
+    ],
+    "upsample_initial_channel": 1536,
+    "resblock": "1",
+    "resblock_kernel_sizes": [
+      3,
+      7,
+      11
+    ],
+    "resblock_dilation_sizes": [
+      [
+        1,
+        3,
+        5
+      ],
+      [
+        1,
+        3,
+        5
+      ],
+      [
+        1,
+        3,
+        5
+      ]
+    ],
+    "downsample_rates": [
+      2,
+      2,
+      2,
+      4,
+      6,
+      10
+    ],
+    "downsample_channels": [
+      12,
+      24,
+      48,
+      96,
+      192,
+      384,
+      768
+    ],
+    "activation": "snakebeta",
+    "snake_logscale": true,
+    "latent_dim": 128,
+    "causal": true,
+    "mi_num_layers": 4,
+    "causal_encoder": true,
+    "use_bias_at_final": false,
+    "use_tanh_at_final": false
+  },
+  "fm_sigma": 0.0,
+  "xvec_drop_rate": 0.2,
+  "campplus_embedding_size": 512,
+  "xvec_max_audio_seconds": 10.0,
+  "architectures": [
+    "DotsTTSForConditionalGeneration"
+  ],
+  "quantization": {
+    "bits": 8,
+    "group_size": 64,
+    "components": [
+      "llm"
+    ]
+  }
+}

models/dots-tts-mlx/int8/core.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c9db4839b276eecd33304bac93823faaa4bae5f527d13efa4e6efb5507e4f77d
+size 2949618031

models/dots-tts-mlx/int8/latent_stats.npz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2fb11a77b710d3230345bd9b0985c075168832ef4b819af317159f31673aa21d
+size 1524

models/dots-tts-mlx/int8/llm_config.json

@@ -0,0 +1,114 @@
+{
+  "vocab_size": 151672,
+  "max_position_embeddings": 131072,
+  "hidden_size": 1536,
+  "intermediate_size": 8960,
+  "num_hidden_layers": 28,
+  "num_attention_heads": 12,
+  "use_sliding_window": false,
+  "sliding_window": null,
+  "max_window_layers": 28,
+  "num_key_value_heads": 2,
+  "hidden_act": "silu",
+  "initializer_range": 0.02,
+  "rms_norm_eps": 1e-06,
+  "use_cache": true,
+  "rope_theta": 1000000.0,
+  "rope_scaling": null,
+  "attention_dropout": 0.0,
+  "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"
+  ],
+  "return_dict": true,
+  "output_hidden_states": false,
+  "torchscript": false,
+  "dtype": "bfloat16",
+  "pruned_heads": {},
+  "tie_word_embeddings": true,
+  "chunk_size_feed_forward": 0,
+  "is_encoder_decoder": false,
+  "is_decoder": false,
+  "cross_attention_hidden_size": null,
+  "add_cross_attention": false,
+  "tie_encoder_decoder": false,
+  "architectures": [
+    "Qwen2ForCausalLM"
+  ],
+  "finetuning_task": null,
+  "id2label": {
+    "0": "LABEL_0",
+    "1": "LABEL_1"
+  },
+  "label2id": {
+    "LABEL_0": 0,
+    "LABEL_1": 1
+  },
+  "task_specific_params": null,
+  "problem_type": null,
+  "tokenizer_class": null,
+  "prefix": null,
+  "bos_token_id": 151643,
+  "pad_token_id": null,
+  "eos_token_id": 151643,
+  "sep_token_id": null,
+  "decoder_start_token_id": null,
+  "max_length": 20,
+  "min_length": 0,
+  "do_sample": false,
+  "early_stopping": false,
+  "num_beams": 1,
+  "temperature": 1.0,
+  "top_k": 50,
+  "top_p": 1.0,
+  "typical_p": 1.0,
+  "repetition_penalty": 1.0,
+  "length_penalty": 1.0,
+  "no_repeat_ngram_size": 0,
+  "encoder_no_repeat_ngram_size": 0,
+  "bad_words_ids": null,
+  "num_return_sequences": 1,
+  "output_scores": false,
+  "return_dict_in_generate": false,
+  "forced_bos_token_id": null,
+  "forced_eos_token_id": null,
+  "remove_invalid_values": false,
+  "exponential_decay_length_penalty": null,
+  "suppress_tokens": null,
+  "begin_suppress_tokens": null,
+  "num_beam_groups": 1,
+  "diversity_penalty": 0.0,
+  "_name_or_path": "",
+  "transformers_version": "4.57.0",
+  "model_type": "qwen2",
+  "use_mrope": false,
+  "tf_legacy_loss": false,
+  "use_bfloat16": false,
+  "output_attentions": false
+}

models/dots-tts-mlx/int8/speaker.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bb418d38516cef3266c7c3cdf3da545f5c32ef1c477d629cc0c8b6a214d855c8
+size 14633169

models/dots-tts-mlx/int8/tokenizer/added_tokens.json

@@ -0,0 +1,31 @@
+{
+  "</tool_call>": 151658,
+  "<tool_call>": 151657,
+  "<|audio_comp_end|>": 151667,
+  "<|audio_comp_span|>": 151666,
+  "<|audio_comp_start|>": 151665,
+  "<|audio_gen_end|>": 151670,
+  "<|audio_gen_span|>": 151669,
+  "<|audio_gen_start|>": 151668,
+  "<|box_end|>": 151649,
+  "<|box_start|>": 151648,
+  "<|endoftext|>": 151643,
+  "<|file_sep|>": 151664,
+  "<|fim_middle|>": 151660,
+  "<|fim_pad|>": 151662,
+  "<|fim_prefix|>": 151659,
+  "<|fim_suffix|>": 151661,
+  "<|im_end|>": 151645,
+  "<|im_start|>": 151644,
+  "<|image_pad|>": 151655,
+  "<|object_ref_end|>": 151647,
+  "<|object_ref_start|>": 151646,
+  "<|quad_end|>": 151651,
+  "<|quad_start|>": 151650,
+  "<|repo_name|>": 151663,
+  "<|text_cond_end|>": 151671,
+  "<|video_pad|>": 151656,
+  "<|vision_end|>": 151653,
+  "<|vision_pad|>": 151654,
+  "<|vision_start|>": 151652
+}

models/dots-tts-mlx/int8/tokenizer/chat_template.jinja

@@ -0,0 +1,54 @@
+{%- if tools %}
+    {{- '<|im_start|>system\n' }}
+    {%- if messages[0]['role'] == 'system' %}
+        {{- messages[0]['content'] }}
+    {%- else %}
+        {{- 'You are a helpful assistant.' }}
+    {%- endif %}
+    {{- "\n\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>" }}
+    {%- for tool in tools %}
+        {{- "\n" }}
+        {{- tool | tojson }}
+    {%- endfor %}
+    {{- "\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" }}
+{%- else %}
+    {%- if messages[0]['role'] == 'system' %}
+        {{- '<|im_start|>system\n' + messages[0]['content'] + '<|im_end|>\n' }}
+    {%- else %}
+        {{- '<|im_start|>system\nYou are a helpful assistant.<|im_end|>\n' }}
+    {%- endif %}
+{%- endif %}
+{%- for message in messages %}
+    {%- if (message.role == "user") or (message.role == "system" and not loop.first) or (message.role == "assistant" and not message.tool_calls) %}
+        {{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>' + '\n' }}
+    {%- elif message.role == "assistant" %}
+        {{- '<|im_start|>' + message.role }}
+        {%- if message.content %}
+            {{- '\n' + message.content }}
+        {%- endif %}
+        {%- for tool_call in message.tool_calls %}
+            {%- if tool_call.function is defined %}
+                {%- set tool_call = tool_call.function %}
+            {%- endif %}
+            {{- '\n<tool_call>\n{"name": "' }}
+            {{- tool_call.name }}
+            {{- '", "arguments": ' }}
+            {{- tool_call.arguments | tojson }}
+            {{- '}\n</tool_call>' }}
+        {%- endfor %}
+        {{- '<|im_end|>\n' }}
+    {%- elif message.role == "tool" %}
+        {%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != "tool") %}
+            {{- '<|im_start|>user' }}
+        {%- endif %}
+        {{- '\n<tool_response>\n' }}
+        {{- message.content }}
+        {{- '\n</tool_response>' }}
+        {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
+            {{- '<|im_end|>\n' }}
+        {%- endif %}
+    {%- endif %}
+{%- endfor %}
+{%- if add_generation_prompt %}
+    {{- '<|im_start|>assistant\n' }}
+{%- endif %}

models/dots-tts-mlx/int8/tokenizer/special_tokens_map.json

@@ -0,0 +1,67 @@
+{
+  "additional_special_tokens": [
+    {
+      "content": "<|audio_comp_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false
+    },
+    {
+      "content": "<|audio_comp_span|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false
+    },
+    {
+      "content": "<|audio_comp_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false
+    },
+    {
+      "content": "<|audio_gen_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false
+    },
+    {
+      "content": "<|audio_gen_span|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false
+    },
+    {
+      "content": "<|audio_gen_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false
+    },
+    {
+      "content": "<|text_cond_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false
+    }
+  ],
+  "eos_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

models/dots-tts-mlx/int8/tokenizer/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c16521f66774c7a4774e5303b7c8ec5c99830c0be5aef6c6edde3ca2a5e05dd0
+size 11423263

models/dots-tts-mlx/int8/tokenizer/tokenizer_config.json

@@ -0,0 +1,257 @@
+{
+  "add_bos_token": false,
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "151643": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151644": {
+      "content": "<|im_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151645": {
+      "content": "<|im_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151646": {
+      "content": "<|object_ref_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151647": {
+      "content": "<|object_ref_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151648": {
+      "content": "<|box_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151649": {
+      "content": "<|box_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151650": {
+      "content": "<|quad_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151651": {
+      "content": "<|quad_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151652": {
+      "content": "<|vision_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151653": {
+      "content": "<|vision_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151654": {
+      "content": "<|vision_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151655": {
+      "content": "<|image_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151656": {
+      "content": "<|video_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151657": {
+      "content": "<tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151658": {
+      "content": "</tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151659": {
+      "content": "<|fim_prefix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151660": {
+      "content": "<|fim_middle|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151661": {
+      "content": "<|fim_suffix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151662": {
+      "content": "<|fim_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151663": {
+      "content": "<|repo_name|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151664": {
+      "content": "<|file_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151665": {
+      "content": "<|audio_comp_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151666": {
+      "content": "<|audio_comp_span|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151667": {
+      "content": "<|audio_comp_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151668": {
+      "content": "<|audio_gen_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151669": {
+      "content": "<|audio_gen_span|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151670": {
+      "content": "<|audio_gen_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151671": {
+      "content": "<|text_cond_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "additional_special_tokens": [
+    "<|audio_comp_start|>",
+    "<|audio_comp_span|>",
+    "<|audio_comp_end|>",
+    "<|audio_gen_start|>",
+    "<|audio_gen_span|>",
+    "<|audio_gen_end|>",
+    "<|text_cond_end|>"
+  ],
+  "bos_token": null,
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|endoftext|>",
+  "errors": "replace",
+  "extra_special_tokens": {},
+  "model_max_length": 131072,
+  "pad_token": "<|endoftext|>",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

models/dots-tts-mlx/int8/vocoder.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9786f810cc674ea13f84c0a163186d6d73ce2c0edbb63b08aaacccd90cb613a1
+size 361798651

models/dots-tts-mlx/languages.txt

@@ -0,0 +1,24 @@
+English
+Chinese
+Yue Chinese
+Japanese
+Korean
+French
+German
+Spanish
+Italian
+Portuguese
+Dutch
+Russian
+Ukrainian
+Polish
+Czech
+Romanian
+Greek
+Finnish
+Turkish
+Arabic
+Hindi
+Vietnamese
+Indonesian
+Thai

models/dots-tts-mlx/mf-int4/config.json

@@ -0,0 +1,127 @@
+{
+  "model_type": "dots_tts",
+  "latent_dim": 128,
+  "patch_size": 4,
+  "cfg_droprate": 0.2,
+  "LLM": {
+    "model_name_or_path": "/prodcpfs/user/jiatu/logs/qwen25_15b"
+  },
+  "PatchEncoder": {
+    "num_layers": 24,
+    "num_heads": 16,
+    "hidden_size": 1024,
+    "ffn_hidden_size": 4096,
+    "modulation": false,
+    "qkv_bias": false,
+    "qk_norm": true,
+    "attn_dropout": 0.0,
+    "dropout": 0.1,
+    "norm_layer": "RMSNorm",
+    "alibi_bias": false,
+    "rotary_bias": true,
+    "rotary_theta": 10000.0,
+    "input_dim": 128,
+    "causal": true
+  },
+  "DiT": {
+    "num_layers": 18,
+    "num_heads": 16,
+    "hidden_size": 1024,
+    "ffn_hidden_size": 4096,
+    "modulation": true,
+    "qkv_bias": false,
+    "qk_norm": true,
+    "attn_dropout": 0.0,
+    "dropout": 0.0,
+    "norm_layer": "RMSNorm",
+    "alibi_bias": false,
+    "rotary_bias": true,
+    "rotary_theta": 10000.0
+  },
+  "vocoder": {
+    "sample_rate": 48000,
+    "upsample_rates": [
+      10,
+      6,
+      4,
+      2,
+      2,
+      2
+    ],
+    "upsample_kernel_sizes": [
+      20,
+      12,
+      8,
+      4,
+      4,
+      4
+    ],
+    "upsample_initial_channel": 1536,
+    "resblock": "1",
+    "resblock_kernel_sizes": [
+      3,
+      7,
+      11
+    ],
+    "resblock_dilation_sizes": [
+      [
+        1,
+        3,
+        5
+      ],
+      [
+        1,
+        3,
+        5
+      ],
+      [
+        1,
+        3,
+        5
+      ]
+    ],
+    "downsample_rates": [
+      2,
+      2,
+      2,
+      4,
+      6,
+      10
+    ],
+    "downsample_channels": [
+      12,
+      24,
+      48,
+      96,
+      192,
+      384,
+      768
+    ],
+    "activation": "snakebeta",
+    "snake_logscale": true,
+    "latent_dim": 128,
+    "causal": true,
+    "mi_num_layers": 4,
+    "causal_encoder": true,
+    "use_bias_at_final": false,
+    "use_tanh_at_final": false
+  },
+  "fm_sigma": 0.0,
+  "xvec_drop_rate": 0.2,
+  "campplus_embedding_size": 512,
+  "xvec_max_audio_seconds": 10.0,
+  "meanflow": {
+    "enabled": true,
+    "use_duration_embedding": true
+  },
+  "architectures": [
+    "DotsTTSForConditionalGeneration"
+  ],
+  "quantization": {
+    "bits": 4,
+    "group_size": 64,
+    "components": [
+      "llm"
+    ]
+  }
+}

models/dots-tts-mlx/mf-int4/core.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:40fdc537eb9982288d585b5a439e95beea7f378dd5d3db36f37b09f5267c39d2
+size 2180661734

models/dots-tts-mlx/mf-int4/latent_stats.npz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2fb11a77b710d3230345bd9b0985c075168832ef4b819af317159f31673aa21d
+size 1524

models/dots-tts-mlx/mf-int4/llm_config.json

@@ -0,0 +1,114 @@
+{
+  "vocab_size": 151672,
+  "max_position_embeddings": 131072,
+  "hidden_size": 1536,
+  "intermediate_size": 8960,
+  "num_hidden_layers": 28,
+  "num_attention_heads": 12,
+  "use_sliding_window": false,
+  "sliding_window": null,
+  "max_window_layers": 28,
+  "num_key_value_heads": 2,
+  "hidden_act": "silu",
+  "initializer_range": 0.02,
+  "rms_norm_eps": 1e-06,
+  "use_cache": true,
+  "rope_theta": 1000000.0,
+  "rope_scaling": null,
+  "attention_dropout": 0.0,
+  "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"
+  ],
+  "return_dict": true,
+  "output_hidden_states": false,
+  "torchscript": false,
+  "dtype": "bfloat16",
+  "pruned_heads": {},
+  "tie_word_embeddings": true,
+  "chunk_size_feed_forward": 0,
+  "is_encoder_decoder": false,
+  "is_decoder": false,
+  "cross_attention_hidden_size": null,
+  "add_cross_attention": false,
+  "tie_encoder_decoder": false,
+  "architectures": [
+    "Qwen2ForCausalLM"
+  ],
+  "finetuning_task": null,
+  "id2label": {
+    "0": "LABEL_0",
+    "1": "LABEL_1"
+  },
+  "label2id": {
+    "LABEL_0": 0,
+    "LABEL_1": 1
+  },
+  "task_specific_params": null,
+  "problem_type": null,
+  "tokenizer_class": null,
+  "prefix": null,
+  "bos_token_id": 151643,
+  "pad_token_id": null,
+  "eos_token_id": 151643,
+  "sep_token_id": null,
+  "decoder_start_token_id": null,
+  "max_length": 20,
+  "min_length": 0,
+  "do_sample": false,
+  "early_stopping": false,
+  "num_beams": 1,
+  "temperature": 1.0,
+  "top_k": 50,
+  "top_p": 1.0,
+  "typical_p": 1.0,
+  "repetition_penalty": 1.0,
+  "length_penalty": 1.0,
+  "no_repeat_ngram_size": 0,
+  "encoder_no_repeat_ngram_size": 0,
+  "bad_words_ids": null,
+  "num_return_sequences": 1,
+  "output_scores": false,
+  "return_dict_in_generate": false,
+  "forced_bos_token_id": null,
+  "forced_eos_token_id": null,
+  "remove_invalid_values": false,
+  "exponential_decay_length_penalty": null,
+  "suppress_tokens": null,
+  "begin_suppress_tokens": null,
+  "num_beam_groups": 1,
+  "diversity_penalty": 0.0,
+  "_name_or_path": "",
+  "transformers_version": "4.57.0",
+  "model_type": "qwen2",
+  "use_mrope": false,
+  "tf_legacy_loss": false,
+  "use_bfloat16": false,
+  "output_attentions": false
+}

models/dots-tts-mlx/mf-int4/speaker.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bb418d38516cef3266c7c3cdf3da545f5c32ef1c477d629cc0c8b6a214d855c8
+size 14633169

models/dots-tts-mlx/mf-int4/tokenizer/added_tokens.json

@@ -0,0 +1,31 @@
+{
+  "</tool_call>": 151658,
+  "<tool_call>": 151657,
+  "<|audio_comp_end|>": 151667,
+  "<|audio_comp_span|>": 151666,
+  "<|audio_comp_start|>": 151665,
+  "<|audio_gen_end|>": 151670,
+  "<|audio_gen_span|>": 151669,
+  "<|audio_gen_start|>": 151668,
+  "<|box_end|>": 151649,
+  "<|box_start|>": 151648,
+  "<|endoftext|>": 151643,
+  "<|file_sep|>": 151664,
+  "<|fim_middle|>": 151660,
+  "<|fim_pad|>": 151662,
+  "<|fim_prefix|>": 151659,
+  "<|fim_suffix|>": 151661,
+  "<|im_end|>": 151645,
+  "<|im_start|>": 151644,
+  "<|image_pad|>": 151655,
+  "<|object_ref_end|>": 151647,
+  "<|object_ref_start|>": 151646,
+  "<|quad_end|>": 151651,
+  "<|quad_start|>": 151650,
+  "<|repo_name|>": 151663,
+  "<|text_cond_end|>": 151671,
+  "<|video_pad|>": 151656,
+  "<|vision_end|>": 151653,
+  "<|vision_pad|>": 151654,
+  "<|vision_start|>": 151652
+}

models/dots-tts-mlx/mf-int4/tokenizer/chat_template.jinja

@@ -0,0 +1,54 @@
+{%- if tools %}
+    {{- '<|im_start|>system\n' }}
+    {%- if messages[0]['role'] == 'system' %}
+        {{- messages[0]['content'] }}
+    {%- else %}
+        {{- 'You are a helpful assistant.' }}
+    {%- endif %}
+    {{- "\n\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>" }}
+    {%- for tool in tools %}
+        {{- "\n" }}
+        {{- tool | tojson }}
+    {%- endfor %}
+    {{- "\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" }}
+{%- else %}
+    {%- if messages[0]['role'] == 'system' %}
+        {{- '<|im_start|>system\n' + messages[0]['content'] + '<|im_end|>\n' }}
+    {%- else %}
+        {{- '<|im_start|>system\nYou are a helpful assistant.<|im_end|>\n' }}
+    {%- endif %}
+{%- endif %}
+{%- for message in messages %}
+    {%- if (message.role == "user") or (message.role == "system" and not loop.first) or (message.role == "assistant" and not message.tool_calls) %}
+        {{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>' + '\n' }}
+    {%- elif message.role == "assistant" %}
+        {{- '<|im_start|>' + message.role }}
+        {%- if message.content %}
+            {{- '\n' + message.content }}
+        {%- endif %}
+        {%- for tool_call in message.tool_calls %}
+            {%- if tool_call.function is defined %}
+                {%- set tool_call = tool_call.function %}
+            {%- endif %}
+            {{- '\n<tool_call>\n{"name": "' }}
+            {{- tool_call.name }}
+            {{- '", "arguments": ' }}
+            {{- tool_call.arguments | tojson }}
+            {{- '}\n</tool_call>' }}
+        {%- endfor %}
+        {{- '<|im_end|>\n' }}
+    {%- elif message.role == "tool" %}
+        {%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != "tool") %}
+            {{- '<|im_start|>user' }}
+        {%- endif %}
+        {{- '\n<tool_response>\n' }}
+        {{- message.content }}
+        {{- '\n</tool_response>' }}
+        {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
+            {{- '<|im_end|>\n' }}
+        {%- endif %}
+    {%- endif %}
+{%- endfor %}
+{%- if add_generation_prompt %}
+    {{- '<|im_start|>assistant\n' }}
+{%- endif %}

models/dots-tts-mlx/mf-int4/tokenizer/special_tokens_map.json

@@ -0,0 +1,67 @@
+{
+  "additional_special_tokens": [
+    {
+      "content": "<|audio_comp_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false
+    },
+    {
+      "content": "<|audio_comp_span|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false
+    },
+    {
+      "content": "<|audio_comp_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false
+    },
+    {
+      "content": "<|audio_gen_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false
+    },
+    {
+      "content": "<|audio_gen_span|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false
+    },
+    {
+      "content": "<|audio_gen_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false
+    },
+    {
+      "content": "<|text_cond_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false
+    }
+  ],
+  "eos_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

models/dots-tts-mlx/mf-int4/tokenizer/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c16521f66774c7a4774e5303b7c8ec5c99830c0be5aef6c6edde3ca2a5e05dd0
+size 11423263

models/dots-tts-mlx/mf-int4/tokenizer/tokenizer_config.json

@@ -0,0 +1,257 @@
+{
+  "add_bos_token": false,
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "151643": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151644": {
+      "content": "<|im_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151645": {
+      "content": "<|im_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151646": {
+      "content": "<|object_ref_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151647": {
+      "content": "<|object_ref_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151648": {
+      "content": "<|box_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151649": {
+      "content": "<|box_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151650": {
+      "content": "<|quad_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151651": {
+      "content": "<|quad_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151652": {
+      "content": "<|vision_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151653": {
+      "content": "<|vision_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151654": {
+      "content": "<|vision_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151655": {
+      "content": "<|image_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151656": {
+      "content": "<|video_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151657": {
+      "content": "<tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151658": {
+      "content": "</tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151659": {
+      "content": "<|fim_prefix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151660": {
+      "content": "<|fim_middle|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151661": {
+      "content": "<|fim_suffix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151662": {
+      "content": "<|fim_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151663": {
+      "content": "<|repo_name|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151664": {
+      "content": "<|file_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151665": {
+      "content": "<|audio_comp_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151666": {
+      "content": "<|audio_comp_span|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151667": {
+      "content": "<|audio_comp_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151668": {
+      "content": "<|audio_gen_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151669": {
+      "content": "<|audio_gen_span|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151670": {
+      "content": "<|audio_gen_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151671": {
+      "content": "<|text_cond_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "additional_special_tokens": [
+    "<|audio_comp_start|>",
+    "<|audio_comp_span|>",
+    "<|audio_comp_end|>",
+    "<|audio_gen_start|>",
+    "<|audio_gen_span|>",
+    "<|audio_gen_end|>",
+    "<|text_cond_end|>"
+  ],
+  "bos_token": null,
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|endoftext|>",
+  "errors": "replace",
+  "extra_special_tokens": {},
+  "model_max_length": 131072,
+  "pad_token": "<|endoftext|>",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

models/dots-tts-mlx/mf-int4/vocoder.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9786f810cc674ea13f84c0a163186d6d73ce2c0edbb63b08aaacccd90cb613a1
+size 361798651