Full-tower republish: 8-bit language tower + BF16 vision tower (fixes mlx_vlm load, discussion #1)

PROVENANCE.md

@@ -0,0 +1,21 @@
+# Provenance
+
+| | |
+|---|---|
+| Upstream | [Qwen/Qwen3.6-35B-A3B](https://huggingface.co/Qwen/Qwen3.6-35B-A3B) @ `995ad96eacd98c81ed38be0c5b274b04031597b0` |
+| Converted | 2026-07-04 |
+| Toolchain | mlx-vlm 0.6.3, mlx 0.31.2, mlx-lm 0.31.3, transformers 5.12.0 |
+| Command | `python -m mlx_vlm convert --hf-path Qwen/Qwen3.6-35B-A3B --revision 995ad96eacd98c81ed38be0c5b274b04031597b0 -q --q-bits 8 --q-group-size 64 --q-mode affine` |
+
+Contents: language tower quantized (8-bit affine, group size 64) +
+**vision tower in BF16** (333 tensors, unquantized — mlx-vlm skips
+multimodal modules by default). The upstream MTP head is not retained
+(dropped by mlx-vlm sanitize; unused at inference).
+
+Smoke (pre-publish, Apple Silicon): `mlx_vlm` vision probe and `mlx_lm`
+text probe both passed for this exact payload.
+
+History: replaces the 2026-04 text-only extraction, which shipped
+without the vision tower and failed to load under `mlx_vlm`
+("Missing 393 parameters"). Reported in discussion #1 on
+majentik/Qwen3.6-35B-A3B-RotorQuant-MLX-4bit.

config.json

@@ -2,10 +2,23 @@
     "architectures": [
         "Qwen3_5MoeForConditionalGeneration"
     ],
+    "do_sample": true,
     "eos_token_id": [
         248046,
         248044
     ],
+    "generation_config": {
+        "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
+    },
     "image_token_id": 248056,
     "model_type": "qwen3_5_moe",
     "quantization": {
@@ -658,6 +671,7 @@
             "bits": 8
         }
     },
+    "temperature": 1.0,
     "text_config": {
         "attention_bias": false,
         "attention_dropout": 0.0,
@@ -741,7 +755,7 @@
             ],
             "partial_rotary_factor": 0.25,
             "rope_theta": 10000000,
-            "type": "default"
+            "rope_type": "default"
         },
         "router_aux_loss_coef": 0.001,
         "shared_expert_intermediate_size": 512,
@@ -750,8 +764,26 @@
         "vocab_size": 248320
     },
     "tie_word_embeddings": false,
+    "top_k": 20,
+    "top_p": 0.95,
     "transformers_version": "4.57.1",
     "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_moe",
+        "num_heads": 16,
+        "num_position_embeddings": 2304,
+        "out_hidden_size": 2048,
+        "patch_size": 16,
+        "spatial_merge_size": 2,
+        "temporal_patch_size": 2
+    },
     "vision_end_token_id": 248054,
     "vision_start_token_id": 248053
 }

configuration.json

@@ -0,0 +1 @@
+{"framework":"Pytorch","task":"visual-question-answering"}

model-00001-of-00007.safetensors → model-00001-of-00008.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c956bf416a5e2176a8f2bafd50a4179992649c234064a23ae7fec0524620f691
-size 5332146307
+oid sha256:143db852a7b5e04ae289a14852079ce5c0465db022ccaeada16629c760a48b81
+size 5329916013

model-00002-of-00007.safetensors → model-00002-of-00008.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c44acb68ee3a36cc1e5ef3b1e8b19157938b3e2604ae8d378ea5c795b2011e08
-size 5358684459
+oid sha256:85ecfd0975ceefee39705bb443a6746613f7ef0e50bb3a11ef9f7d89755a0ca7
+size 5365573922

model-00003-of-00007.safetensors → model-00003-of-00008.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:455f4ea92763f5ec18ffac372bd5dda7b5c7884e021bd9f1d19c6f4ffb48980d
-size 5365574144
+oid sha256:939da359a8bb9e1ce190c0b0b94b0d62d44522ab0755117cf86ae228d823a7ae
+size 5358684668

model-00004-of-00008.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e2f66bfcd864a385073f0e4b4c1a9daf57e858bddd9f291769c14cc22c9c1cce
+size 5365574194

model-00005-of-00007.safetensors

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:da2b620a957e41a7ea3ff590d45111a01e6fc97c0ad32d6c20961699bb63645a
-size 5365574154

model-00004-of-00007.safetensors → model-00005-of-00008.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2f89972fd32bad9959e0bac8b980fdf8742c3890eb057d61f2d72820ad655a24
+oid sha256:6d669425cfaa312e65a65f6b7ee0fad208534640483e33fd592c91f5fb5334a4
 size 5358684644

model-00006-of-00007.safetensors

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:aac2ae720118e0be63cbacc475afb4e6c6e5fa80ccef00def44daed4bc0563a3
-size 5358684646

model-00006-of-00008.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:29c219880920fc1880f30eea25a41b98b251bda318e1f958d4df2bbd74e6e2b0
+size 5365574174

model-00007-of-00007.safetensors

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:0c41d2487a0c839754a694f09bfec16bb8acb6815ae82876633f6b2d8a74c448
-size 4688874171

model-00007-of-00008.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d231a4e95d408f4fa780d61d0011f1cf7843cff12602e68356714e821bd7390e
+size 5037050574

model-00008-of-00008.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:145818a7e8f32071c56f6f817aa452aff235745f82033f141f07f5f6f7bc41c6
+size 540344667

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 +1,50 @@
+{
+  "image_processor": {
+    "do_convert_rgb": true,
+    "do_normalize": true,
+    "do_rescale": true,
+    "image_mean": [
+      0.5,
+      0.5,
+      0.5
+    ],
+    "image_processor_type": "Qwen3VLImageProcessor",
+    "image_std": [
+      0.5,
+      0.5,
+      0.5
+    ],
+    "max_pixels": 16777216,
+    "merge_size": 2,
+    "min_pixels": 65536,
+    "patch_size": 16,
+    "rescale_factor": 0.00392156862745098,
+    "temporal_patch_size": 2
+  },
+  "processor_class": "Qwen3VLProcessor",
+  "video_processor": {
+    "do_convert_rgb": true,
+    "do_normalize": true,
+    "do_rescale": true,
+    "fps": 2.0,
+    "image_mean": [
+      0.5,
+      0.5,
+      0.5
+    ],
+    "image_std": [
+      0.5,
+      0.5,
+      0.5
+    ],
+    "max_frames": 768,
+    "max_pixels": 25165824,
+    "merge_size": 2,
+    "min_frames": 4,
+    "min_pixels": 4096,
+    "patch_size": 16,
+    "rescale_factor": 0.00392156862745098,
+    "temporal_patch_size": 2,
+    "video_processor_type": "Qwen3VLVideoProcessor"
+  }
+}

tokenizer.json

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:87a7830d63fcf43bf241c3c5242e96e62dd3fdc29224ca26fed8ea333db72de4
-size 19989343
+oid sha256:06b9509352d2af50381ab2247e083b80d32d5c0aba91c272ca9ff729b6a0e523
+size 19989325

tokenizer_config.json

@@ -10,6 +10,7 @@
   "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|>",
@@ -22,9 +23,9 @@
   },
   "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+",
+  "processor_class": "Qwen3VLProcessor",
   "split_special_tokens": false,
-  "tokenizer_class": "TokenizersBackend",
-  "tool_parser_type": "qwen3_coder",
+  "tokenizer_class": "Qwen2Tokenizer",
   "unk_token": null,
   "video_token": "<|video_pad|>",
   "vision_bos_token": "<|vision_start|>",

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"
+}