Duplicate from Youssofal/Qwen3.8-27B-MTPLX-Optimized-Quality

Co-authored-by: Youssof Altoukhi <Youssofal@users.noreply.huggingface.co>

.gitattributes

@@ -0,0 +1,36 @@
+*.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
+tokenizer.json filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,123 @@
+---
+license: apache-2.0
+library_name: mtplx
+pipeline_tag: text-generation
+base_model: Qwen/Qwen3.8-27B
+base_model_relation: quantized
+tags:
+- mlx
+- apple-silicon
+- macos
+- speculative-decoding
+- multi-token-prediction
+- qwen
+- qwen3.8
+- mtp
+- mtplx
+- local-ai
+- coding
+- qwen3-8
+- qwen-3.8
+- local-llm
+- llm
+- m5
+- m5-max
+- m4
+- m3
+- macbook-pro
+- mac-studio
+- opencode
+- claude-code
+- 27b
+- qwen3.8-27b
+- qwen3-8-27b
+---
+
+**[MTPLX](https://mtplx.com): the fastest way to run Qwen 3.8 on a Mac. Native multi-token-prediction speculative decoding on Apple Silicon, two to three times the speed of plain decoding, exact at any temperature.**
+
+# Qwen 3.8 27B Optimized Quality
+
+**8-bit dynamic quant. Good coding speeds and perfect quality.**
+
+The closest of the three MTPLX Qwen 3.8 builds to the original bf16 model.
+Every weight matrix at 8-bit, sensitive parts at 16-bit, native
+multi-token-prediction head kept, so [MTPLX](https://mtplx.com) still drafts
+ahead and verifies in one pass. Pick this when you want the answer the full
+model would give and still want it fast.
+
+## Measured on MTPLX 2.11.3 (16 September 2026)
+
+This is a Qwen3.8-27B MLX pack for MTPLX, the fastest way to run Qwen 3.8 27B on a Mac. All numbers on a MacBook Pro M5 Max, fans verified at maximum, sampled at the model's own settings. Conditions and sources: [mtplx.com/benchmarks](https://mtplx.com/benchmarks/).
+
+| Run | tok/s |
+|---|---|
+| Optimized Speed rewriting a file it just wrote, stock settings (MTPLX 2.10.0) | 87.6 |
+| Bare Speed on a fresh coding task at official Qwen 3.8 sampling (MTPLX 2.7.0) | 65.2 |
+| Optimized Speed on a 3k-token chat answer (MTPLX 2.10.0) | 64.3 |
+| The 27B record on a fresh generation: Qwen 3.6 27B Optimized Speed, 192-token bench, 2 July 2026, raw logs published | 81.74 |
+
+Pack quality against the bf16 Qwen3.8-27B checkpoint, teacher-forced over 2,389 positions of code, prose, JSON and a multilingual notice (16 September 2026): Optimized Speed (4-bit dynamic) 96.0 percent top-1 agreement and KL 0.012; Optimized Quality (8-bit dynamic) 99.3 percent and KL 0.0005. Exactness on this release: a thousand four-token draws from the fast path match a thousand from the plain path within the plain path's own noise, at temperature 1, top-p 0.95, top-k 20. Details: [MTPLX 2.11.3 release notes](https://mtplx.com/releases/2.11.3/).
+
+Runs on Apple Silicon Macs with 32 GB of unified memory or more (36 GB for Optimized Quality): MacBook Pro, MacBook Air, Mac mini and Mac Studio on M1 to M5. Guide: [Run Qwen 3.8 27B on a Mac](https://mtplx.com/models/qwen3.8-27b/). Comparison pages: [MTPLX vs mlx-serve](https://mtplx.com/compare/mtplx-vs-mlx-serve/), [MTPLX vs oMLX](https://mtplx.com/compare/mtplx-vs-omlx/), [MTPLX vs LM Studio](https://mtplx.com/compare/mtplx-vs-lm-studio/), [MTPLX vs Ollama](https://mtplx.com/compare/mtplx-vs-ollama/).
+
+## Speeds
+
+Measured on an M5 Max, fans verified at max, single stream, generation running
+to the model's own stop, official Qwen 3.8 sampling (temperature 1.0, top-p
+0.95, top-k 20).
+
+| Run | tok/s |
+|---|---|
+| Coding task, medium reasoning, inside the MTPLX Mac app | 48.3 |
+| Long reasoning at xhigh, 34k and 46k token answers | 33.2 and 33.1 |
+
+Same night, same task, the 4-bit builds: Qwen 3.6 27B Optimized Speed V2 59.9
+to 60.1 tok/s, Qwen 3.8 Optimized Speed 58.7, Bare Speed 65.2. This is the
+quality pick, not the speed pick, and it is still well past 40 tok/s while
+running the full-precision distribution.
+
+Draft acceptance on the coding task by depth: 0.96, 0.88, 0.79. Verify cost
+63.5 ms per round. Depth 3 was measured at +19.9% over depth 2 on the long
+reasoning task.
+
+## How it is built
+
+- Every weight matrix at 8-bit with 64-weight groups.
+- The GDN convolution kernels and recurrent state parameters, every norm, and
+  the whole MTP head stay 16-bit.
+- KL divergence to the original bf16 model on our coding battery: 0.00105.
+  That is 21x closer than Optimized Speed and 36x closer than Bare Speed. In
+  practice you will not tell the outputs apart from the bf16 model.
+
+| | |
+|---|---|
+| Download | 29.4 GB |
+| Peak unified memory (measured, this artifact) | 32.7 GB |
+| Context window | 262,144 tokens |
+| MTP depth | 3 |
+| Sampling | temperature 1.0, top-p 0.95, top-k 20 (the official Qwen 3.8 contract) |
+
+The tuned depth and draft settings ship inside `mtplx_runtime.json`. MTPLX
+reads them on load. Speculation is exact: drafts are accepted with the
+probability-ratio rule plus residual resampling, so the output follows the
+model's own distribution at any temperature. Reasoning effort levels (xhigh,
+medium, low) work, and preserved thinking flows through the MTP path.
+
+## Use it
+
+You want 36 GB of unified memory or more for this one. Mac app: download at
+[mtplx.com](https://mtplx.com), pick "Qwen 3.8 27B Optimized Quality".
+
+Command line:
+
+```bash
+pip install mtplx
+mtplx serve --model Youssofal/Qwen3.8-27B-MTPLX-Optimized-Quality
+```
+
+Siblings: [Optimized Speed](https://huggingface.co/Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed)
+(recommended for coding) and
+[Bare Speed](https://huggingface.co/Youssofal/Qwen3.8-27B-MTPLX-Bare-Speed)
+(quickest burst chat speeds). On an M1 or M2 Mac use the
+[FP16 build](https://huggingface.co/Youssofal/Qwen3.8-27B-MTPLX-Optimized-Quality-FP16)
+of this model.

chat_template.jinja

@@ -0,0 +1,170 @@
+{%- set image_count = namespace(value=0) %}
+{%- set video_count = namespace(value=0) %}
+{%- macro render_content(content, do_vision_count, is_system_content=false) %}
+    {%- if content is string %}
+        {{- content }}
+    {%- elif content is iterable and content is not mapping %}
+        {%- for item in content %}
+            {%- if 'image' in item or 'image_url' in item or item.type == 'image' %}
+                {%- if is_system_content %}
+                    {{- raise_exception('System message cannot contain images.') }}
+                {%- endif %}
+                {%- if do_vision_count %}
+                    {%- set image_count.value = image_count.value + 1 %}
+                {%- endif %}
+                {%- if add_vision_id %}
+                    {{- 'Picture ' ~ image_count.value ~ ': ' }}
+                {%- endif %}
+                {{- '<|vision_start|><|image_pad|><|vision_end|>' }}
+            {%- elif 'video' in item or item.type == 'video' %}
+                {%- if is_system_content %}
+                    {{- raise_exception('System message cannot contain videos.') }}
+                {%- endif %}
+                {%- if do_vision_count %}
+                    {%- set video_count.value = video_count.value + 1 %}
+                {%- endif %}
+                {%- if add_vision_id %}
+                    {{- 'Video ' ~ video_count.value ~ ': ' }}
+                {%- endif %}
+                {{- '<|vision_start|><|video_pad|><|vision_end|>' }}
+            {%- elif 'text' in item %}
+                {{- item.text }}
+            {%- else %}
+                {{- raise_exception('Unexpected item type in content.') }}
+            {%- endif %}
+        {%- endfor %}
+    {%- elif content is none or content is undefined %}
+        {{- '' }}
+    {%- else %}
+        {{- raise_exception('Unexpected content type.') }}
+    {%- endif %}
+{%- endmacro %}
+{%- if not messages %}
+    {{- raise_exception('No messages provided.') }}
+{%- endif %}
+{%- set reasoning_instructions = '' %}
+{%- if enable_thinking is undefined or enable_thinking is true %}
+    {%- set resolved_reasoning_effort = reasoning_effort|default('xhigh') %}
+    {%- if resolved_reasoning_effort not in ('xhigh', 'medium', 'low') %}
+        {{- raise_exception('Unexpected reasoning effort ' ~ reasoning_effort ~ '. Supported types are xhigh (default), medium, and low.') }}
+    {%- endif %}
+    {%- if resolved_reasoning_effort == 'xhigh' %}
+        {%- set reasoning_instructions = 'Reasoning effort is set to xhigh. Please think carefully through the task, validate key assumptions, consider plausible alternatives, and prioritize correctness, consistency, and clarity in the final answer.' %}
+    {%- elif resolved_reasoning_effort == 'low' %}
+        {%- set reasoning_instructions = 'Reasoning effort is set to low. Keep your thinking brief and focused, moving directly to the conclusion without unnecessary elaboration.' %}
+    {%- endif %}
+{%- endif %}
+{%- if tools and tools is iterable and tools is not mapping %}
+    {{- '<|im_start|>system\n' }}
+    {%- if reasoning_instructions %}
+        {{- reasoning_instructions + '\n\n' }}
+    {%- endif %}
+    {{- "# Tools\n\nYou have access to the following functions:\n\n<tools>" }}
+    {%- for tool in tools %}
+        {{- "\n" }}
+        {{- tool | tojson }}
+    {%- endfor %}
+    {{- "\n</tools>" }}
+    {{- '\n\nIf you choose to call a function ONLY reply in the following format with NO suffix:\n\n<tool_call>\n<function=example_function_name>\n<parameter=example_parameter_1>\nvalue_1\n</parameter>\n<parameter=example_parameter_2>\nThis is the value for the second parameter\nthat can span\nmultiple lines\n</parameter>\n</function>\n</tool_call>\n\n<IMPORTANT>\nReminder:\n- Function calls MUST follow the specified format: an inner <function=...></function> block must be nested within <tool_call></tool_call> XML tags\n- Required parameters MUST be specified\n- You may provide optional reasoning for your function call in natural language BEFORE the function call, but NOT after\n- If there is no function call available, answer the question like normal with your current knowledge and do not tell the user about function calls\n</IMPORTANT>' }}
+    {%- if messages[0].role == 'system' %}
+        {%- set content = render_content(messages[0].content, false, true)|trim %}
+        {%- if content %}
+            {{- '\n\n' + content }}
+        {%- endif %}
+    {%- endif %}
+    {{- '<|im_end|>\n' }}
+{%- else %}
+    {%- if messages[0].role == 'system' %}
+        {%- set content = render_content(messages[0].content, false, true)|trim %}
+        {%- if content %}
+            {{- '<|im_start|>system\n' + (reasoning_instructions + '\n\n' if reasoning_instructions else '')  + content + '<|im_end|>\n' }}
+        {%- elif reasoning_instructions %}
+            {{- '<|im_start|>system\n' + reasoning_instructions + '<|im_end|>\n' }}
+        {%- endif %}
+    {%- elif reasoning_instructions %}
+        {{- '<|im_start|>system\n' + reasoning_instructions + '<|im_end|>\n' }}
+    {%- endif %}
+{%- endif %}
+{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
+{%- for message in messages[::-1] %}
+    {%- set index = (messages|length - 1) - loop.index0 %}
+    {%- if ns.multi_step_tool and message.role == "user" %}
+        {%- set content = render_content(message.content, false)|trim %}
+        {%- if not(content.startswith('<tool_response>') and content.endswith('</tool_response>')) %}
+            {%- set ns.multi_step_tool = false %}
+            {%- set ns.last_query_index = index %}
+        {%- endif %}
+    {%- endif %}
+{%- endfor %}
+{%- if ns.multi_step_tool %}
+    {{- raise_exception('No user query found in messages.') }}
+{%- endif %}
+{%- for message in messages %}
+    {%- set content = render_content(message.content, true)|trim %}
+    {%- if message.role == "system" %}
+        {%- if not loop.first %}
+            {{- raise_exception('System message must be at the beginning.') }}
+        {%- endif %}
+    {%- elif message.role == "user" %}
+        {{- '<|im_start|>' + message.role + '\n' + content + '<|im_end|>' + '\n' }}
+    {%- elif message.role == "assistant" %}
+        {%- set reasoning_content = '' %}
+        {%- if message.reasoning_content is string %}
+            {%- set reasoning_content = message.reasoning_content %}
+        {%- endif %}
+        {%- set reasoning_content = reasoning_content|trim %}
+        {%- if preserve_thinking is undefined or preserve_thinking is true or loop.index0 > ns.last_query_index %}
+            {{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content + '\n</think>\n\n' + content }}
+        {%- else %}
+            {{- '<|im_start|>' + message.role + '\n' + content }}
+        {%- endif %}
+        {%- if message.tool_calls and message.tool_calls is iterable and message.tool_calls is not mapping %}
+            {%- for tool_call in message.tool_calls %}
+                {%- if tool_call.function is defined %}
+                    {%- set tool_call = tool_call.function %}
+                {%- endif %}
+                {%- if loop.first %}
+                    {%- if content|trim %}
+                        {{- '\n\n<tool_call>\n<function=' + tool_call.name + '>\n' }}
+                    {%- else %}
+                        {{- '<tool_call>\n<function=' + tool_call.name + '>\n' }}
+                    {%- endif %}
+                {%- else %}
+                    {{- '\n<tool_call>\n<function=' + tool_call.name + '>\n' }}
+                {%- endif %}
+                {%- if tool_call.arguments is defined and tool_call.arguments != '' %}
+                    {%- for args_name, args_value in tool_call.arguments|items %}
+                        {{- '<parameter=' + args_name + '>\n' }}
+                        {%- set args_value = args_value | string if args_value is string else args_value | tojson | safe %}
+                        {{- args_value }}
+                        {{- '\n</parameter>\n' }}
+                    {%- endfor %}
+                {%- endif %}
+                {{- '</function>\n</tool_call>' }}
+            {%- endfor %}
+        {%- endif %}
+        {{- '<|im_end|>\n' }}
+    {%- elif message.role == "tool" %}
+        {%- if loop.previtem and loop.previtem.role != "tool" %}
+            {{- '<|im_start|>user' }}
+        {%- endif %}
+        {{- '\n<tool_response>\n' }}
+        {{- content }}
+        {{- '\n</tool_response>' }}
+        {%- if not loop.last and loop.nextitem.role != "tool" %}
+            {{- '<|im_end|>\n' }}
+        {%- elif loop.last %}
+            {{- '<|im_end|>\n' }}
+        {%- endif %}
+    {%- else %}
+        {{- raise_exception('Unexpected message role.') }}
+    {%- endif %}
+{%- endfor %}
+{%- if add_generation_prompt %}
+    {{- '<|im_start|>assistant\n' }}
+    {%- if enable_thinking is defined and enable_thinking is false %}
+        {{- '<think>\n\n</think>\n\n' }}
+    {%- else %}
+        {{- '<think>\n' }}
+    {%- endif %}
+{%- endif %}

config.json

@@ -0,0 +1,184 @@
+{
+  "architectures": [
+    "Qwen3_5ForConditionalGeneration"
+  ],
+  "eos_token_id": [
+    248046,
+    248044
+  ],
+  "image_token_id": 248056,
+  "language_model_only": false,
+  "mlx_lm_extra_tensors": {
+    "mtp_file": "mtp.safetensors"
+  },
+  "model_type": "qwen3_5",
+  "mtplx_mtp_contract": {
+    "base_hidden_variant": "post_norm",
+    "concat_order": "embedding_hidden",
+    "hidden_variant": "post_norm",
+    "mtp_position_mode": "local",
+    "mtp_quant_group_size": 64,
+    "mtp_quant_mode": "affine"
+  },
+  "mtplx_mtp_payload_audit": {
+    "mtp_file": "/Users/youssof/.mtplx/models/Qwen3.8-27B-MTPLX-Optimized-Quality/mtp.safetensors",
+    "nonzero_payload_tensor_count": 8,
+    "passed": true,
+    "payload_tensor_count": 8,
+    "problems": [],
+    "scale_tensor_count": 0,
+    "tensor_count": 15,
+    "zero_payload_sample": [],
+    "zero_scale_sample": []
+  },
+  "quantization": {
+    "bits": 8,
+    "group_size": 64,
+    "mode": "affine"
+  },
+  "quantization_config": {
+    "bits": 8,
+    "group_size": 64,
+    "mode": "affine"
+  },
+  "text_config": {
+    "attention_bias": false,
+    "attention_dropout": 0.0,
+    "attn_output_gate": true,
+    "bos_token_id": 248044,
+    "dtype": "bfloat16",
+    "eos_token_id": 248044,
+    "full_attention_interval": 4,
+    "head_dim": 256,
+    "hidden_act": "silu",
+    "hidden_size": 5120,
+    "initializer_range": 0.02,
+    "intermediate_size": 17408,
+    "layer_types": [
+      "linear_attention",
+      "linear_attention",
+      "linear_attention",
+      "full_attention",
+      "linear_attention",
+      "linear_attention",
+      "linear_attention",
+      "full_attention",
+      "linear_attention",
+      "linear_attention",
+      "linear_attention",
+      "full_attention",
+      "linear_attention",
+      "linear_attention",
+      "linear_attention",
+      "full_attention",
+      "linear_attention",
+      "linear_attention",
+      "linear_attention",
+      "full_attention",
+      "linear_attention",
+      "linear_attention",
+      "linear_attention",
+      "full_attention",
+      "linear_attention",
+      "linear_attention",
+      "linear_attention",
+      "full_attention",
+      "linear_attention",
+      "linear_attention",
+      "linear_attention",
+      "full_attention",
+      "linear_attention",
+      "linear_attention",
+      "linear_attention",
+      "full_attention",
+      "linear_attention",
+      "linear_attention",
+      "linear_attention",
+      "full_attention",
+      "linear_attention",
+      "linear_attention",
+      "linear_attention",
+      "full_attention",
+      "linear_attention",
+      "linear_attention",
+      "linear_attention",
+      "full_attention",
+      "linear_attention",
+      "linear_attention",
+      "linear_attention",
+      "full_attention",
+      "linear_attention",
+      "linear_attention",
+      "linear_attention",
+      "full_attention",
+      "linear_attention",
+      "linear_attention",
+      "linear_attention",
+      "full_attention",
+      "linear_attention",
+      "linear_attention",
+      "linear_attention",
+      "full_attention"
+    ],
+    "linear_conv_kernel_dim": 4,
+    "linear_key_head_dim": 128,
+    "linear_num_key_heads": 16,
+    "linear_num_value_heads": 48,
+    "linear_value_head_dim": 128,
+    "mamba_ssm_dtype": "float32",
+    "max_position_embeddings": 262144,
+    "model_type": "qwen3_5_text",
+    "mtp_num_hidden_layers": 1,
+    "mtp_use_dedicated_embeddings": false,
+    "num_attention_heads": 24,
+    "num_hidden_layers": 64,
+    "num_key_value_heads": 4,
+    "output_gate_type": "swish",
+    "pad_token_id": null,
+    "partial_rotary_factor": 0.25,
+    "rms_norm_eps": 1e-06,
+    "rope_parameters": {
+      "mrope_interleaved": true,
+      "mrope_section": [
+        11,
+        11,
+        10
+      ],
+      "partial_rotary_factor": 0.25,
+      "rope_theta": 10000000,
+      "type": "default"
+    },
+    "tie_word_embeddings": false,
+    "use_cache": true,
+    "vocab_size": 248320
+  },
+  "tie_word_embeddings": false,
+  "transformers_version": "5.8.0.dev0",
+  "video_token_id": 248057,
+  "vision_config": {
+    "deepstack_visual_indexes": [],
+    "depth": 27,
+    "hidden_act": "gelu_pytorch_tanh",
+    "hidden_size": 1152,
+    "in_channels": 3,
+    "initializer_range": 0.02,
+    "intermediate_size": 4304,
+    "model_type": "qwen3_5",
+    "num_heads": 16,
+    "num_position_embeddings": 2304,
+    "out_hidden_size": 5120,
+    "patch_size": 16,
+    "spatial_merge_size": 2,
+    "temporal_patch_size": 2
+  },
+  "vision_end_token_id": 248054,
+  "vision_start_token_id": 248053,
+  "mtplx_mtp_quantization": {
+    "bits": 8,
+    "group_size": 64,
+    "mode": "affine",
+    "policy": "all",
+    "prequantized": true,
+    "description": "All 8 MTP draft-head matrices (fc + attention q/k/v/o + MLP gate/up/down) packed MLX INT8/g64 affine from the released sidecar; head norms keep the pack's float dtype. Verified flat-or-better acceptance vs the unquantized head before publishing."
+  }
+}

generation_config.json

@@ -0,0 +1,12 @@
+{
+    "bos_token_id": 248044,
+    "do_sample": true,
+    "eos_token_id": [
+        248046,
+        248044
+    ],
+    "pad_token_id": 248044,
+    "temperature": 1.0,
+    "top_k": 20,
+    "top_p": 0.95
+}

model-00001-of-00006.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a11adc73146b142e77349999f67d5d13c031ff2bbafed0837388c1b6e48d4836
+size 5305469313

model-00002-of-00006.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bd95f804980c5e79e212abe6e674020a30f24674780a92cb101dd9971484c6d0
+size 5354102628

model-00003-of-00006.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:91e3f7cc9fadd3599ebce87a9f8e8a4ab6d05f595a79da94868fa96c2d868b8e
+size 5337392986

model-00004-of-00006.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ae6fee0e32edbf973e614ac040ad0f853df23f8c33bf97cadb4c168ef9a887a2
+size 5292847128

model-00005-of-00006.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c1735ac9c3d6cba8127b5c00a28576f0825d874a7f09911dc7bad5027a8747eb
+size 5354102635

model-00006-of-00006.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2452a44eddf64d0cd964245b37f47f3b406c9c397b0eb5fd605bfb9c2f8825f9
+size 1935805825

model-vision.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:964bef26c740bdb6fe464b4c7c48840d4c952ea597c181c41cb131c65ea3c5d5
+size 921497225

mtp.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1e350715e0509932e337eb60c074a3e47be1a9a04ea0fcb6c29b193b67c470c5
+size 451270903

mtplx_runtime.json

@@ -0,0 +1,179 @@
+{
+  "arch_id": "qwen3-next-mtp",
+  "artifact_role": "forge-local",
+  "base_trunk": "Qwen/Qwen3.8-27B",
+  "exactness_baseline": {},
+  "forge_provenance": {
+    "depth_default_note": "Depth pin REMOVED 2026-08-14 after the gated live ABBA: D2 39.5 vs D3 40.6 tok/s matched-window (tie; D3 fewer verify rounds 697 vs 883). The earlier forge-verify \"D3 18.8 collapse\" was order/JIT-confounded (mistakes/single-forge-verify-tune-rows...). Family ceiling D3 applies.",
+    "forge_inputs": {
+      "mtp_source_path": "<redacted>/Qwen--Qwen3.8-27B",
+      "trunk_path": "<redacted>/Qwen3.8-27B-MTPLX-Optimized-Quality"
+    },
+    "forge_recipe": {
+      "body_bits": 8,
+      "body_group_size": 64,
+      "body_mode": "affine",
+      "mtp_policy": "keep_bf16"
+    },
+    "forged_at": "2026-08-14T11:23:31-07:00",
+    "forged_locally": true,
+    "head_quantization": {
+      "bits": 8,
+      "group_size": 64,
+      "mode": "affine",
+      "note": "Structural head quantization of the released sidecar; no calibration, no training. Trunk weights unchanged.",
+      "policy": "all",
+      "quantized_at": "2026-08-20T04:49:56Z",
+      "quantized_sidecar_bytes": 451270903,
+      "source_sidecar_bytes": 849400403,
+      "tool": "scripts/build_qwen38_q4head_sidecar.py"
+    },
+    "mtp_contract": {
+      "base_hidden_variant": "post_norm",
+      "concat_order": "embedding_hidden",
+      "hidden_variant": "post_norm",
+      "mtp_position_mode": "local",
+      "mtp_quant_group_size": 64,
+      "mtp_quant_mode": "affine"
+    },
+    "mtplx_version": "2.6.0",
+    "published_to_hf": null,
+    "recommended_profile_note": "Restamped sustained->turbo 2026-08-14: the sustained recommendation came from the same confounded verify rows; every gated arm served turbo, medium Flappy 59.8 tok/s blended quiet-window.",
+    "source_format": "bf16_native",
+    "source_repo": "Qwen/Qwen3.8-27B",
+    "source_sha": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0"
+  },
+  "mtp_contract": {
+    "base_hidden_variant": "post_norm",
+    "concat_order": "embedding_hidden",
+    "hidden_variant": "post_norm",
+    "mtp_position_mode": "local",
+    "mtp_quant_group_size": 64,
+    "mtp_quant_mode": "affine"
+  },
+  "mtp_depth_default": 3,
+  "mtp_depth_max": 3,
+  "mtp_sidecar": "int8-g64-prequantized",
+  "mtplx_version": "2.9.0",
+  "recommended_draft_sampler": {
+    "temperature": 1.0,
+    "top_k": 20,
+    "top_p": 0.95
+  },
+  "recommended_profile": "turbo",
+  "release_validation": {
+    "draft_sampler_transfer_note": {
+      "basis": "dropday uncapped ABBA campaign (clean-receipts.jsonl, temp 1.0 xhigh/medium, die-gated, max fans verified)",
+      "note": "draft 1.0 adopted by transfer: OS uncapped ABBA (+13%) + Bare scaffold strict A/B (keep_official_draft_sampler_1_0); same MTP head architecture and temp-1.0 target contract. OQ depth arms measured at draft 0.6; app-receipt validates live at ship config.",
+      "stamped_at": "2026-08-14T21:45:00-07:00"
+    },
+    "dropday_depth_abba": {
+      "basis": "dropday uncapped ABBA campaign (clean-receipts.jsonl, temp 1.0 xhigh/medium, die-gated, max fans verified)",
+      "decode_tps_blended_d2": [
+        27.3,
+        28.0
+      ],
+      "decode_tps_blended_d3": [
+        33.2,
+        33.1
+      ],
+      "stamped_at": "2026-08-14T21:45:00-07:00",
+      "verdict": "depth_3 (+19.9% blended)",
+      "workload": "simple flappy xhigh uncapped (33k-46k tokens, EOS stop)"
+    }
+  },
+  "sampler": {
+    "temperature": 0.6,
+    "top_k": 20,
+    "top_p": 0.95
+  },
+  "speed_evidence": {
+    "acceptance_by_depth": [
+      0.9433962264150944,
+      0.8814016172506739,
+      0.8059299191374663
+    ],
+    "acceptance_collapsed": [],
+    "artifact_fingerprint": "sha256:5a7a0d400ebff91dc544f7d93454d6d48d2833b4f0f7ad832d00e1d9513fd750",
+    "depth": 3,
+    "failure_reasons": [],
+    "forge_verify_rows": [
+      {
+        "acceptance_by_position": [],
+        "depth": 0,
+        "finish_reasons": {
+          "stop": 1
+        },
+        "hit_token_budget": false,
+        "hit_token_budget_count": 0,
+        "multiplier_vs_ar": 1.0,
+        "quality_passed": true,
+        "tok_s": 13.054796369549772,
+        "verify_time_s": 154.4356367919827
+      },
+      {
+        "acceptance_by_position": [
+          0.9570405727923628
+        ],
+        "depth": 1,
+        "finish_reasons": {
+          "stop": 1
+        },
+        "hit_token_budget": false,
+        "hit_token_budget_count": 0,
+        "multiplier_vs_ar": 2.0262953935160954,
+        "quality_passed": true,
+        "tok_s": 26.45287374690935,
+        "verify_time_s": 35.17482435004786
+      },
+      {
+        "acceptance_by_position": [
+          0.9776119402985075,
+          0.9129353233830846
+        ],
+        "depth": 2,
+        "finish_reasons": {
+          "stop": 1
+        },
+        "hit_token_budget": false,
+        "hit_token_budget_count": 0,
+        "multiplier_vs_ar": 2.560816141204823,
+        "quality_passed": true,
+        "tok_s": 33.430933263285176,
+        "verify_time_s": 36.15608408872504
+      },
+      {
+        "acceptance_by_position": [
+          0.9433962264150944,
+          0.8814016172506739,
+          0.8059299191374663
+        ],
+        "depth": 3,
+        "finish_reasons": {
+          "stop": 1
+        },
+        "hit_token_budget": false,
+        "hit_token_budget_count": 0,
+        "multiplier_vs_ar": 3.002168329361034,
+        "quality_passed": true,
+        "tok_s": 39.192696206919734,
+        "verify_time_s": 31.090109037584625
+      }
+    ],
+    "greedy_diagnostic": {
+      "tok_s": 13.054796369549772
+    },
+    "quality_rejected": [],
+    "tok_s": [
+      39.192696206919734
+    ],
+    "verdict": "mtp_depth_wins"
+  },
+  "verified_on": {
+    "hardware": "macOS-26.3.1-arm64-arm-64bit-Mach-O",
+    "machine_arch": "arm64",
+    "macos": "26.3.1",
+    "model": "Qwen3.8-27B-MTPLX-Optimized-Quality",
+    "timestamp": "2026-08-19T23:08:02-07:00"
+  }
+}

preprocessor_config.json

@@ -0,0 +1,21 @@
+{
+    "size": {
+        "longest_edge": 16777216,
+        "shortest_edge": 65536
+    },
+    "patch_size": 16,
+    "temporal_patch_size": 2,
+    "merge_size": 2,
+    "image_mean": [
+        0.5,
+        0.5,
+        0.5
+    ],
+    "image_std": [
+        0.5,
+        0.5,
+        0.5
+    ],
+    "processor_class": "Qwen3VLProcessor",
+    "image_processor_type": "Qwen2VLImageProcessorFast"
+}

processor_config.json

@@ -0,0 +1,17 @@
+{
+  "image_processor": {
+    "image_processor_type": "Qwen2VLImageProcessor",
+    "size": {
+      "longest_edge": 16777216,
+      "shortest_edge": 65536
+    }
+  },
+  "processor_class": "Qwen3VLProcessor",
+  "video_processor": {
+    "size": {
+      "longest_edge": 25165824,
+      "shortest_edge": 4096
+    },
+    "video_processor_type": "Qwen3VLVideoProcessor"
+  }
+}

tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:06b9509352d2af50381ab2247e083b80d32d5c0aba91c272ca9ff729b6a0e523
+size 19989325

tokenizer_config.json

@@ -0,0 +1,33 @@
+{
+  "add_prefix_space": false,
+  "audio_bos_token": "<|audio_start|>",
+  "audio_eos_token": "<|audio_end|>",
+  "audio_token": "<|audio_pad|>",
+  "backend": "tokenizers",
+  "bos_token": null,
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "errors": "replace",
+  "image_token": "<|image_pad|>",
+  "is_local": true,
+  "local_files_only": false,
+  "model_max_length": 262144,
+  "model_specific_special_tokens": {
+    "audio_bos_token": "<|audio_start|>",
+    "audio_eos_token": "<|audio_end|>",
+    "audio_token": "<|audio_pad|>",
+    "image_token": "<|image_pad|>",
+    "video_token": "<|video_pad|>",
+    "vision_bos_token": "<|vision_start|>",
+    "vision_eos_token": "<|vision_end|>"
+  },
+  "pad_token": "<|endoftext|>",
+  "pretokenize_regex": "(?i:'s|'t|'re|'ve|'m|'ll|'d)|[^\\r\\n\\p{L}\\p{N}]?[\\p{L}\\p{M}]+|\\p{N}| ?[^\\s\\p{L}\\p{M}\\p{N}]+[\\r\\n]*|\\s*[\\r\\n]+|\\s+(?!\\S)|\\s+",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "tool_parser_type": "qwen3_coder",
+  "unk_token": null,
+  "video_token": "<|video_pad|>",
+  "vision_bos_token": "<|vision_start|>",
+  "vision_eos_token": "<|vision_end|>"
+}

video_preprocessor_config.json

@@ -0,0 +1,21 @@
+{
+    "size": {
+        "longest_edge": 25165824,
+        "shortest_edge": 4096
+    },
+    "patch_size": 16,
+    "temporal_patch_size": 2,
+    "merge_size": 2,
+    "image_mean": [
+        0.5,
+        0.5,
+        0.5
+    ],
+    "image_std": [
+        0.5,
+        0.5,
+        0.5
+    ],
+    "processor_class": "Qwen3VLProcessor",
+    "video_processor_type": "Qwen3VLVideoProcessor"
+}