Hugging Face's logo

shadowlilac
/
MiMo-V2.5-NVFP4

Safetensors
mimo_v2
custom_code
8-bit precision
modelopt
Model card Files
xet
 Community
shadowlilac lukealonso commited on
Commit
ea4b70d
·
0 Parent(s):

Duplicate from lukealonso/MiMo-V2.5-NVFP4

Browse files

Co-authored-by: Luke Alonso <lukealonso@users.noreply.huggingface.co>

This view is limited to 50 files because it contains too many changes.   See raw diff
Files changed (50) hide show
  1. .gitattributes +36 -0
  2. README.md +83 -0
  3. amax_checkpoint.safetensors +3 -0
  4. audio_tokenizer/chat_template.jinja +120 -0
  5. audio_tokenizer/config.json +68 -0
  6. audio_tokenizer/generation_config.json +9 -0
  7. audio_tokenizer/model.safetensors +3 -0
  8. audio_tokenizer/tokenizer_config.json +267 -0
  9. config.json +598 -0
  10. generation_config.json +9 -0
  11. hf_quant_config.json +227 -0
  12. merges.txt +0 -0
  13. model-00001-of-00035.safetensors +3 -0
  14. model-00002-of-00035.safetensors +3 -0
  15. model-00003-of-00035.safetensors +3 -0
  16. model-00004-of-00035.safetensors +3 -0
  17. model-00005-of-00035.safetensors +3 -0
  18. model-00006-of-00035.safetensors +3 -0
  19. model-00007-of-00035.safetensors +3 -0
  20. model-00008-of-00035.safetensors +3 -0
  21. model-00009-of-00035.safetensors +3 -0
  22. model-00010-of-00035.safetensors +3 -0
  23. model-00011-of-00035.safetensors +3 -0
  24. model-00012-of-00035.safetensors +3 -0
  25. model-00013-of-00035.safetensors +3 -0
  26. model-00014-of-00035.safetensors +3 -0
  27. model-00015-of-00035.safetensors +3 -0
  28. model-00016-of-00035.safetensors +3 -0
  29. model-00017-of-00035.safetensors +3 -0
  30. model-00018-of-00035.safetensors +3 -0
  31. model-00019-of-00035.safetensors +3 -0
  32. model-00020-of-00035.safetensors +3 -0
  33. model-00021-of-00035.safetensors +3 -0
  34. model-00022-of-00035.safetensors +3 -0
  35. model-00023-of-00035.safetensors +3 -0
  36. model-00024-of-00035.safetensors +3 -0
  37. model-00025-of-00035.safetensors +3 -0
  38. model-00026-of-00035.safetensors +3 -0
  39. model-00027-of-00035.safetensors +3 -0
  40. model-00028-of-00035.safetensors +3 -0
  41. model-00029-of-00035.safetensors +3 -0
  42. model-00030-of-00035.safetensors +3 -0
  43. model-00031-of-00035.safetensors +3 -0
  44. model-00032-of-00035.safetensors +3 -0
  45. model-00033-of-00035.safetensors +3 -0
  46. model-00034-of-00035.safetensors +3 -0
  47. model-00035-of-00035.safetensors +3 -0
  48. model-inputscales.safetensors +3 -0
  49. model-mtp.safetensors +3 -0
  50. model.safetensors.index.json +3 -0
.gitattributes ADDED
@@ -0,0 +1,36 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ *.7z filter=lfs diff=lfs merge=lfs -text
2
+ *.arrow filter=lfs diff=lfs merge=lfs -text
3
+ *.bin filter=lfs diff=lfs merge=lfs -text
4
+ *.bz2 filter=lfs diff=lfs merge=lfs -text
5
+ *.ckpt filter=lfs diff=lfs merge=lfs -text
6
+ *.ftz filter=lfs diff=lfs merge=lfs -text
7
+ *.gz filter=lfs diff=lfs merge=lfs -text
8
+ *.h5 filter=lfs diff=lfs merge=lfs -text
9
+ *.joblib filter=lfs diff=lfs merge=lfs -text
10
+ *.lfs.* filter=lfs diff=lfs merge=lfs -text
11
+ *.mlmodel filter=lfs diff=lfs merge=lfs -text
12
+ *.model filter=lfs diff=lfs merge=lfs -text
13
+ *.msgpack filter=lfs diff=lfs merge=lfs -text
14
+ *.npy filter=lfs diff=lfs merge=lfs -text
15
+ *.npz filter=lfs diff=lfs merge=lfs -text
16
+ *.onnx filter=lfs diff=lfs merge=lfs -text
17
+ *.ot filter=lfs diff=lfs merge=lfs -text
18
+ *.parquet filter=lfs diff=lfs merge=lfs -text
19
+ *.pb filter=lfs diff=lfs merge=lfs -text
20
+ *.pickle filter=lfs diff=lfs merge=lfs -text
21
+ *.pkl filter=lfs diff=lfs merge=lfs -text
22
+ *.pt filter=lfs diff=lfs merge=lfs -text
23
+ *.pth filter=lfs diff=lfs merge=lfs -text
24
+ *.rar filter=lfs diff=lfs merge=lfs -text
25
+ *.safetensors filter=lfs diff=lfs merge=lfs -text
26
+ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
27
+ *.tar.* filter=lfs diff=lfs merge=lfs -text
28
+ *.tar filter=lfs diff=lfs merge=lfs -text
29
+ *.tflite filter=lfs diff=lfs merge=lfs -text
30
+ *.tgz filter=lfs diff=lfs merge=lfs -text
31
+ *.wasm filter=lfs diff=lfs merge=lfs -text
32
+ *.xz filter=lfs diff=lfs merge=lfs -text
33
+ *.zip filter=lfs diff=lfs merge=lfs -text
34
+ *.zst filter=lfs diff=lfs merge=lfs -text
35
+ *tfevents* filter=lfs diff=lfs merge=lfs -text
36
+ model.safetensors.index.json filter=lfs diff=lfs merge=lfs -text
README.md ADDED
@@ -0,0 +1,83 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model:
3
+ - XiaomiMiMo/MiMo-V2.5
4
+ ---
5
+
6
+ ## IMPORTANT: You *must* use the docker image below, since it contains many custom kernels written for this model specifically ##
7
+
8
+ ## Model Description
9
+
10
+ **MiMo-V2.5-NVFP4** is an NVFP4-quantized version of [XiaomiMiMo/MiMo-V2.5](https://huggingface.co/XiaomiMiMo/MiMo-V2.5).
11
+
12
+ This is a multi-modal model, supporting text, images, audio and video. This quantization carefully preserves those capabilities.
13
+
14
+ ### What's quantized
15
+
16
+ Only the *non-shared* MoE expert MLP projections are quantized to NVFP4. Attention weights are left in BF16, in addition to the dense MLPs (layers 0-3) and the shared experts. Since the MoE expert weights constitute the vast majority of model parameters in an MoE architecture, this still yields significant memory savings.
17
+
18
+ Calibration uses natural top-k routing rather than forcing all experts to activate, so each expert's quantization scales reflect the token distributions it actually sees during inference. To compensate, calibration was run on a much larger number of samples than typical to ensure broad expert coverage through natural routing alone.
19
+
20
+ ### Calibration dataset
21
+
22
+ Six calibration passes were run:
23
+
24
+ 1. **Coding** — Agentic coding samples (tool calling, multi-turn code generation, function calling) with English and Chinese system prompts.
25
+ 2. **Broad** — Large-scale diverse samples drawn from WildChat-NonToxic and LMSYS-Chat covering real user conversations across a wide range of topics and languages.
26
+ 3. **Deep** — Long-context samples (>8K tokens) from coding and diverse sources to exercise deep-sequence expert activation patterns.
27
+ 4. **Image** — Image question-answering prompts, with the input images drawn from a large collection of public, high quality image datasets.
28
+ 5. **Audio** — Medium-size dataset of mostly speech.
29
+ 6. **Video** — Diverse set of video question-answering prompts, with a wide variety of input videos of different durations and resolutions.
30
+
31
+ ### Requirements
32
+
33
+ The NVFP4 variant of this model is currently only supported on RTX 6000 (SM120), due to the large number of custom kernels that had to be written to support it.
34
+
35
+ Minimum: 2x RTX PRO 6000 Blackwell 96GB (future memory optimizations forthcoming that'll allow it to fit better, for now you'll have to the model sequence length, batch size to make it fit)
36
+
37
+ Recommended: 4x RTX PRO 6000 Blackwell 96GB
38
+
39
+ ### Community Testing
40
+
41
+ Note: You will of course want to modify this to bind mount your HF cache, or you'll re-download the model each time.
42
+
43
+ ```
44
+ docker run --rm -it \
45
+ --name sglang-mimo-v25 \
46
+ --gpus '"device=0,1,2,3"' \
47
+ --ipc=host \
48
+ --network host \
49
+ --ulimit memlock=-1 \
50
+ --ulimit stack=67108864 \
51
+ -e OMP_NUM_THREADS=16 \
52
+ -e SAFETENSORS_FAST_GPU=1 \
53
+ -e CUTE_DSL_ARCH="sm_120a" \
54
+ docker.io/lukealonso/sglang-cuda13-b12x \
55
+ python -m sglang.launch_server \
56
+ --model-path lukealonso/MiMo-V2.5-NVFP4 \
57
+ --served-model-name "MiMo-V2.5" \
58
+ --tp-size 4 \
59
+ --page-size 64 \
60
+ --host 0.0.0.0 \
61
+ --port 8000 \
62
+ --enforce-piecewise-cuda-graph \
63
+ --kv-cache-dtype fp8_e4m3 \
64
+ --mem-fraction-static 0.85 \
65
+ --chunked-prefill-size 8192 \
66
+ --speculative-algorithm EAGLE \
67
+ --speculative-num-steps 3 \
68
+ --speculative-eagle-topk 1 \
69
+ --speculative-num-draft-tokens 4 \
70
+ --enable-pcie-oneshot-allreduce \
71
+ --enable-multi-layer-eagle \
72
+ --reasoning-parser mimo \
73
+ --tool-call-parser mimo \
74
+ --quantization modelopt_fp4 \
75
+ --max-running-requests 8 \
76
+ --moe-runner-backend b12x \
77
+ --attention-backend b12x \
78
+ --mm-attention-backend b12x \
79
+ --fp4-gemm-backend b12x
80
+
81
+ ```
82
+
83
+
amax_checkpoint.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8242c5179df4c3ece758b0f86d26078693bb10baeec88ca79c5ea2a0e375148c
3
+ size 8619248
audio_tokenizer/chat_template.jinja ADDED
@@ -0,0 +1,120 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {%- if tools %}
2
+ {{- '<|im_start|>system\n' }}
3
+ {%- if messages[0].role == 'system' %}
4
+ {%- if messages[0].content is string %}
5
+ {{- messages[0].content }}
6
+ {%- else %}
7
+ {%- for content in messages[0].content %}
8
+ {%- if content.type == 'audio' %}
9
+ {{- ("<|sosp|>" + (content.meta | tojson) + "<|eosp|>") }}
10
+ {%- elif content.type == 'text' %}
11
+ {{- content.text }}
12
+ {%- endif %}
13
+ {%- endfor %}
14
+ {%- endif %}
15
+ {%- endif %}
16
+ {{- '\n\n' }}
17
+ {{- "# 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>" }}
18
+ {%- for tool in tools %}
19
+ {{- "\n" }}
20
+ {{- tool | tojson }}
21
+ {%- endfor %}
22
+ {{- "\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" }}
23
+ {%- else %}
24
+ {%- if messages[0].role == 'system' %}
25
+ {{- '<|im_start|>system\n' }}
26
+ {%- if messages[0].content is string %}
27
+ {{- messages[0].content }}
28
+ {%- else %}
29
+ {%- for content in messages[0].content %}
30
+ {%- if content.type == 'audio' %}
31
+ {{- ("<|sosp|>" + (content.meta | tojson) + "<|eosp|>") }}
32
+ {%- elif content.type == 'text' %}
33
+ {{- content.text }}
34
+ {%- endif %}
35
+ {%- endfor %}
36
+ {%- endif %}
37
+ {{- '\n<|im_end|>\n' }}
38
+ {%- endif %}
39
+ {%- endif %}
40
+ {%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1, assistant_is_last=false) %}
41
+ {%- for message in messages[::-1] %}
42
+ {%- set index = (messages|length - 1) - loop.index0 %}
43
+ {%- if ns.multi_step_tool and message.role == "user" and message.content is string and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}
44
+ {%- set ns.multi_step_tool = false %}
45
+ {%- set ns.last_query_index = index %}
46
+ {%- endif %}
47
+ {%- endfor %}
48
+ {%- for message in messages %}
49
+ {%- if message.content is string %}
50
+ {%- set content = message.content %}
51
+ {%- else %}
52
+ {%- set content = namespace(text="") %}
53
+ {%- for mcontent in message.content %}
54
+ {%- if mcontent.type == 'audio' %}
55
+ {%- set content.text = content.text~("<|sosp|>" + (mcontent.meta | tojson) + "<|eosp|>") %}
56
+ {%- elif mcontent.type == 'text' %}
57
+ {%- set content.text = content.text~mcontent.text %}
58
+ {%- endif %}
59
+ {%- endfor %}
60
+ {%- set content = content.text %}
61
+ {%- endif %}
62
+ {%- if (message.role == "user") or (message.role == "system" and not loop.first) %}
63
+ {{- '<|im_start|>' + message.role + '\n' + content + '<|im_end|>' + '\n' }}
64
+ {%- elif message.role == "assistant" %}
65
+ {%- set reasoning_content = "" %}
66
+ {%- if message.reasoning_content is string %}
67
+ {%- set reasoning_content = message.reasoning_content %}
68
+ {%- else %}
69
+ {%- if '</think>' in content %}
70
+ {%- set reasoning_content = content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
71
+ {%- set content = content.split('</think>')[-1].lstrip('\n') %}
72
+ {%- endif %}
73
+ {%- endif %}
74
+ {%- if loop.index0 > ns.last_query_index %}
75
+ {%- if loop.last or (not loop.last and reasoning_content) %}
76
+ {{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content.strip("\n") + '\n</think>\n\n' + content.lstrip('\n') }}
77
+ {%- else %}
78
+ {{- '<|im_start|>' + message.role + '\n' + content }}
79
+ {%- endif %}
80
+ {%- else %}
81
+ {{- '<|im_start|>' + message.role + '\n' + content }}
82
+ {%- endif %}
83
+ {%- if message.tool_calls %}
84
+ {%- for tool_call in message.tool_calls %}
85
+ {%- if (loop.first and content) or (not loop.first) %}{{- '\n' }}{%- endif %}
86
+ {%- if tool_call.function %}
87
+ {%- set tool_call = tool_call.function %}
88
+ {%- endif %}
89
+ {{- '<tool_call>\n{"name": "' }}
90
+ {{- tool_call.name }}
91
+ {{- '", "arguments": ' }}
92
+ {%- if tool_call.arguments is string %}
93
+ {{- tool_call.arguments }}
94
+ {%- else %}
95
+ {{- tool_call.arguments | tojson }}
96
+ {%- endif %}
97
+ {{- '}\n</tool_call>' }}
98
+ {%- endfor %}
99
+ {%- endif %}
100
+ {%- if loop.last %}
101
+ {%- set ns.assistant_is_last = true %}
102
+ {%- else %}
103
+ {{- '<|im_end|>\n' }}
104
+ {%- endif %}
105
+ {%- elif message.role == "tool" %}
106
+ {%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}{{- '<|im_start|>user' }}{%- endif %}
107
+ {{- '\n<tool_response>\n' }}
108
+ {{- content }}
109
+ {{- '\n</tool_response>' }}
110
+ {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}{{- '<|im_end|>\n' }}{%- endif %}
111
+ {%- endif %}
112
+ {%- endfor %}
113
+ {%- if add_generation_prompt and not ns.assistant_is_last %}
114
+ {{- '<|im_start|>assistant\n' }}
115
+ {%- if audio_output %}
116
+ {{- '<|sostm|>'}}
117
+ {%- elif not enable_thinking %}
118
+ {{- '<think>\n\n</think>\n' }}
119
+ {%- endif %}
120
+ {%- endif %}
audio_tokenizer/config.json ADDED
@@ -0,0 +1,68 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "max_audio_seconds": 300,
3
+ "stride_size": 2,
4
+ "avg_pooler": 2,
5
+ "d_model": 1024,
6
+ "scale_embedding": false,
7
+ "kernel_size": 3,
8
+ "activation_function": "gelu",
9
+ "encoder_layers": 24,
10
+ "encoder_skip_layer_id": 3,
11
+ "encoder_attention_heads": 16,
12
+ "encoder_ffn_dim": 4096,
13
+ "encoder_causal": true,
14
+ "encoder_attn_window_size": [
15
+ 128,
16
+ 0
17
+ ],
18
+ "decoder_layers": 24,
19
+ "decoder_attention_heads": 16,
20
+ "decoder_ffn_dim": 4096,
21
+ "decoder_kernel_size": 3,
22
+ "decoder_stride_size": 2,
23
+ "decoder_causal": true,
24
+ "decoder_attn_window_size": [
25
+ 128,
26
+ 0
27
+ ],
28
+ "nfft": 960,
29
+ "n_mels": 128,
30
+ "sampling_rate": 24000,
31
+ "hop_length": 240,
32
+ "window_size": 960,
33
+ "vocoder_padding": "same",
34
+ "fmin": 0,
35
+ "fmax": null,
36
+ "num_quantizers": 20,
37
+ "codebook_size": [
38
+ 1024,
39
+ 1024,
40
+ 256,
41
+ 128,
42
+ 128,
43
+ 128,
44
+ 128,
45
+ 128,
46
+ 128,
47
+ 128,
48
+ 128,
49
+ 128,
50
+ 128,
51
+ 128,
52
+ 128,
53
+ 128,
54
+ 128,
55
+ 128,
56
+ 128,
57
+ 128
58
+ ],
59
+ "threshold_ema_dead_code": 2,
60
+ "position_embedding_type": "rope",
61
+ "rope_theta": 10000,
62
+ "rope_type": "default",
63
+ "ln_type": "LayerNorm",
64
+ "use_istft_only": true,
65
+ "hybrid_attention": true,
66
+ "hybrid_block_size": 8,
67
+ "swa_per_block": 2
68
+ }
audio_tokenizer/generation_config.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "do_sample": true,
3
+ "temperature": 0.6,
4
+ "top_k": -1,
5
+ "top_p": 0.95,
6
+ "audio_temperature": 0.9,
7
+ "audio_top_k": -1,
8
+ "audio_top_p": 0.95
9
+ }
audio_tokenizer/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:95cca046bda0a67ea52cc77af734ed175282820efbc508099dd8a012eb968cea
3
+ size 652622472
audio_tokenizer/tokenizer_config.json ADDED
@@ -0,0 +1,267 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": false,
3
+ "add_prefix_space": false,
4
+ "added_tokens_decoder": {
5
+ "151643": {
6
+ "content": "<|endoftext|>",
7
+ "lstrip": false,
8
+ "normalized": false,
9
+ "rstrip": false,
10
+ "single_word": false,
11
+ "special": true
12
+ },
13
+ "151644": {
14
+ "content": "<|im_start|>",
15
+ "lstrip": false,
16
+ "normalized": false,
17
+ "rstrip": false,
18
+ "single_word": false,
19
+ "special": true
20
+ },
21
+ "151645": {
22
+ "content": "<|im_end|>",
23
+ "lstrip": false,
24
+ "normalized": false,
25
+ "rstrip": false,
26
+ "single_word": false,
27
+ "special": true
28
+ },
29
+ "151646": {
30
+ "content": "<|object_ref_start|>",
31
+ "lstrip": false,
32
+ "normalized": false,
33
+ "rstrip": false,
34
+ "single_word": false,
35
+ "special": true
36
+ },
37
+ "151647": {
38
+ "content": "<|object_ref_end|>",
39
+ "lstrip": false,
40
+ "normalized": false,
41
+ "rstrip": false,
42
+ "single_word": false,
43
+ "special": true
44
+ },
45
+ "151648": {
46
+ "content": "<|box_start|>",
47
+ "lstrip": false,
48
+ "normalized": false,
49
+ "rstrip": false,
50
+ "single_word": false,
51
+ "special": true
52
+ },
53
+ "151649": {
54
+ "content": "<|box_end|>",
55
+ "lstrip": false,
56
+ "normalized": false,
57
+ "rstrip": false,
58
+ "single_word": false,
59
+ "special": true
60
+ },
61
+ "151650": {
62
+ "content": "<|quad_start|>",
63
+ "lstrip": false,
64
+ "normalized": false,
65
+ "rstrip": false,
66
+ "single_word": false,
67
+ "special": true
68
+ },
69
+ "151651": {
70
+ "content": "<|quad_end|>",
71
+ "lstrip": false,
72
+ "normalized": false,
73
+ "rstrip": false,
74
+ "single_word": false,
75
+ "special": true
76
+ },
77
+ "151652": {
78
+ "content": "<|vision_start|>",
79
+ "lstrip": false,
80
+ "normalized": false,
81
+ "rstrip": false,
82
+ "single_word": false,
83
+ "special": true
84
+ },
85
+ "151653": {
86
+ "content": "<|vision_end|>",
87
+ "lstrip": false,
88
+ "normalized": false,
89
+ "rstrip": false,
90
+ "single_word": false,
91
+ "special": true
92
+ },
93
+ "151654": {
94
+ "content": "<|vision_pad|>",
95
+ "lstrip": false,
96
+ "normalized": false,
97
+ "rstrip": false,
98
+ "single_word": false,
99
+ "special": true
100
+ },
101
+ "151655": {
102
+ "content": "<|image_pad|>",
103
+ "lstrip": false,
104
+ "normalized": false,
105
+ "rstrip": false,
106
+ "single_word": false,
107
+ "special": true
108
+ },
109
+ "151656": {
110
+ "content": "<|video_pad|>",
111
+ "lstrip": false,
112
+ "normalized": false,
113
+ "rstrip": false,
114
+ "single_word": false,
115
+ "special": true
116
+ },
117
+ "151657": {
118
+ "content": "<tool_call>",
119
+ "lstrip": false,
120
+ "normalized": false,
121
+ "rstrip": false,
122
+ "single_word": false,
123
+ "special": false
124
+ },
125
+ "151658": {
126
+ "content": "</tool_call>",
127
+ "lstrip": false,
128
+ "normalized": false,
129
+ "rstrip": false,
130
+ "single_word": false,
131
+ "special": false
132
+ },
133
+ "151659": {
134
+ "content": "<|fim_prefix|>",
135
+ "lstrip": false,
136
+ "normalized": false,
137
+ "rstrip": false,
138
+ "single_word": false,
139
+ "special": false
140
+ },
141
+ "151660": {
142
+ "content": "<|fim_middle|>",
143
+ "lstrip": false,
144
+ "normalized": false,
145
+ "rstrip": false,
146
+ "single_word": false,
147
+ "special": false
148
+ },
149
+ "151661": {
150
+ "content": "<|fim_suffix|>",
151
+ "lstrip": false,
152
+ "normalized": false,
153
+ "rstrip": false,
154
+ "single_word": false,
155
+ "special": false
156
+ },
157
+ "151662": {
158
+ "content": "<|fim_pad|>",
159
+ "lstrip": false,
160
+ "normalized": false,
161
+ "rstrip": false,
162
+ "single_word": false,
163
+ "special": false
164
+ },
165
+ "151663": {
166
+ "content": "<|repo_name|>",
167
+ "lstrip": false,
168
+ "normalized": false,
169
+ "rstrip": false,
170
+ "single_word": false,
171
+ "special": false
172
+ },
173
+ "151664": {
174
+ "content": "<|file_sep|>",
175
+ "lstrip": false,
176
+ "normalized": false,
177
+ "rstrip": false,
178
+ "single_word": false,
179
+ "special": false
180
+ },
181
+ "151665": {
182
+ "content": "<|mimo_audio_start|>",
183
+ "lstrip": false,
184
+ "normalized": false,
185
+ "rstrip": false,
186
+ "single_word": false,
187
+ "special": true
188
+ },
189
+ "151666": {
190
+ "content": "<|mimo_audio_end|>",
191
+ "lstrip": false,
192
+ "normalized": false,
193
+ "rstrip": false,
194
+ "single_word": false,
195
+ "special": true
196
+ },
197
+ "151667": {
198
+ "content": "<think>",
199
+ "lstrip": false,
200
+ "normalized": false,
201
+ "rstrip": false,
202
+ "single_word": false,
203
+ "special": false
204
+ },
205
+ "151668": {
206
+ "content": "</think>",
207
+ "lstrip": false,
208
+ "normalized": false,
209
+ "rstrip": false,
210
+ "single_word": false,
211
+ "special": false
212
+ },
213
+ "151669": {
214
+ "content": "<|audio_pad|>",
215
+ "lstrip": false,
216
+ "normalized": false,
217
+ "rstrip": false,
218
+ "single_word": false,
219
+ "special": true
220
+ },
221
+ "151670": {
222
+ "content": "<|mimo_video_start|>",
223
+ "lstrip": false,
224
+ "normalized": false,
225
+ "rstrip": false,
226
+ "single_word": false,
227
+ "special": true
228
+ },
229
+ "151671": {
230
+ "content": "<|mimo_video_end|>",
231
+ "lstrip": false,
232
+ "normalized": false,
233
+ "rstrip": false,
234
+ "single_word": false,
235
+ "special": true
236
+ }
237
+ },
238
+ "additional_special_tokens": [
239
+ "<|im_start|>",
240
+ "<|im_end|>",
241
+ "<|object_ref_start|>",
242
+ "<|object_ref_end|>",
243
+ "<|box_start|>",
244
+ "<|box_end|>",
245
+ "<|quad_start|>",
246
+ "<|quad_end|>",
247
+ "<|vision_start|>",
248
+ "<|vision_end|>",
249
+ "<|vision_pad|>",
250
+ "<|image_pad|>",
251
+ "<|video_pad|>",
252
+ "<|audio_pad|>",
253
+ "<|mimo_audio_start|>",
254
+ "<|mimo_audio_end|>",
255
+ "<|mimo_video_start|>",
256
+ "<|mimo_video_end|>"
257
+ ],
258
+ "bos_token": null,
259
+ "clean_up_tokenization_spaces": false,
260
+ "eos_token": "<|im_end|>",
261
+ "errors": "replace",
262
+ "model_max_length": 131072,
263
+ "pad_token": "<|endoftext|>",
264
+ "split_special_tokens": false,
265
+ "tokenizer_class": "Qwen2Tokenizer",
266
+ "unk_token": null
267
+ }
config.json ADDED
@@ -0,0 +1,598 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "attention_projection_layout": "fused_qkv",
3
+ "vocab_size": 152576,
4
+ "max_position_embeddings": 1048576,
5
+ "hidden_size": 4096,
6
+ "intermediate_size": 16384,
7
+ "num_hidden_layers": 48,
8
+ "num_attention_heads": 64,
9
+ "num_key_value_heads": 4,
10
+ "hidden_act": "silu",
11
+ "initializer_range": 0.02,
12
+ "layernorm_epsilon": 1e-05,
13
+ "use_cache": true,
14
+ "rope_theta": 10000000,
15
+ "rope_parameters": {
16
+ "rope_type": "default",
17
+ "type": "default",
18
+ "rope_theta": 10000000,
19
+ "partial_rotary_factor": 0.334
20
+ },
21
+ "attention_dropout": 0.0,
22
+ "attention_bias": false,
23
+ "attention_value_scale": 0.707,
24
+ "head_dim": 192,
25
+ "v_head_dim": 128,
26
+ "swa_num_attention_heads": 64,
27
+ "swa_num_key_value_heads": 8,
28
+ "swa_head_dim": 192,
29
+ "swa_v_head_dim": 128,
30
+ "swa_rope_theta": 10000,
31
+ "sliding_window": 128,
32
+ "sliding_window_size": 128,
33
+ "add_full_attention_sink_bias": false,
34
+ "add_swa_attention_sink_bias": true,
35
+ "hybrid_block_size": null,
36
+ "hybrid_layer_pattern": [
37
+ 0,
38
+ 1,
39
+ 1,
40
+ 1,
41
+ 1,
42
+ 0,
43
+ 1,
44
+ 1,
45
+ 1,
46
+ 1,
47
+ 1,
48
+ 0,
49
+ 1,
50
+ 1,
51
+ 1,
52
+ 1,
53
+ 1,
54
+ 0,
55
+ 1,
56
+ 1,
57
+ 1,
58
+ 1,
59
+ 1,
60
+ 0,
61
+ 1,
62
+ 1,
63
+ 1,
64
+ 1,
65
+ 1,
66
+ 0,
67
+ 1,
68
+ 1,
69
+ 1,
70
+ 1,
71
+ 1,
72
+ 0,
73
+ 1,
74
+ 1,
75
+ 1,
76
+ 1,
77
+ 1,
78
+ 0,
79
+ 1,
80
+ 1,
81
+ 1,
82
+ 1,
83
+ 1,
84
+ 0
85
+ ],
86
+ "partial_rotary_factor": 0.334,
87
+ "n_routed_experts": 256,
88
+ "moe_intermediate_size": 2048,
89
+ "num_experts_per_tok": 8,
90
+ "routed_scaling_factor": null,
91
+ "scoring_func": "sigmoid",
92
+ "topk_method": "noaux_tc",
93
+ "n_group": 1,
94
+ "topk_group": 1,
95
+ "norm_topk_prob": true,
96
+ "moe_layer_freq": [
97
+ 0,
98
+ 1,
99
+ 1,
100
+ 1,
101
+ 1,
102
+ 1,
103
+ 1,
104
+ 1,
105
+ 1,
106
+ 1,
107
+ 1,
108
+ 1,
109
+ 1,
110
+ 1,
111
+ 1,
112
+ 1,
113
+ 1,
114
+ 1,
115
+ 1,
116
+ 1,
117
+ 1,
118
+ 1,
119
+ 1,
120
+ 1,
121
+ 1,
122
+ 1,
123
+ 1,
124
+ 1,
125
+ 1,
126
+ 1,
127
+ 1,
128
+ 1,
129
+ 1,
130
+ 1,
131
+ 1,
132
+ 1,
133
+ 1,
134
+ 1,
135
+ 1,
136
+ 1,
137
+ 1,
138
+ 1,
139
+ 1,
140
+ 1,
141
+ 1,
142
+ 1,
143
+ 1,
144
+ 1
145
+ ],
146
+ "vision_config": {
147
+ "depth": 28,
148
+ "fullatt_block_indexes": [
149
+ 0,
150
+ 9,
151
+ 18,
152
+ 27
153
+ ],
154
+ "hidden_act": "silu",
155
+ "hidden_size": 1280,
156
+ "in_chans": 3,
157
+ "intermediate_size": 4608,
158
+ "num_heads": 32,
159
+ "num_key_value_heads": 8,
160
+ "num_query_groups": 4,
161
+ "out_hidden_size": 4096,
162
+ "patch_size": 16,
163
+ "spatial_merge_size": 2,
164
+ "spatial_patch_size": 16,
165
+ "temporal_patch_size": 2,
166
+ "tokens_per_second": 2,
167
+ "use_sink": true,
168
+ "visual_token_window_size": 64,
169
+ "vit_window_attn_types": [
170
+ -1,
171
+ 0,
172
+ 0,
173
+ 0,
174
+ 0,
175
+ 1,
176
+ 1,
177
+ 1,
178
+ 1,
179
+ -1,
180
+ 0,
181
+ 0,
182
+ 0,
183
+ 0,
184
+ 1,
185
+ 1,
186
+ 1,
187
+ 1,
188
+ -1,
189
+ 0,
190
+ 0,
191
+ 0,
192
+ 0,
193
+ 1,
194
+ 1,
195
+ 1,
196
+ 1,
197
+ -1
198
+ ],
199
+ "window_size": 128
200
+ },
201
+ "audio_config": {
202
+ "add_post_norm": true,
203
+ "audio_channels": 20,
204
+ "audio_segment_size": 6000,
205
+ "group_size": 4,
206
+ "input_full_attention": true,
207
+ "input_local_attn_heads": 16,
208
+ "input_local_dim": 1024,
209
+ "input_local_head_dim": 64,
210
+ "input_local_hidden_dropout": 0.0,
211
+ "input_local_intermediate_size": 4096,
212
+ "input_local_layers": 6,
213
+ "out_hidden_size": 4096,
214
+ "partial_rotary_factor": 1.0,
215
+ "projection_layers": 2,
216
+ "rope_theta": 640000,
217
+ "speech_vocab_size": "1280",
218
+ "speech_zeroemb_idx": "1024"
219
+ },
220
+ "processor_config": {
221
+ "audio_avg_pooler": 2,
222
+ "audio_channels": 20,
223
+ "audio_end_token_id": 151674,
224
+ "audio_fmax": null,
225
+ "audio_fmin": 0,
226
+ "audio_group_size": 4,
227
+ "audio_hop_length": 240,
228
+ "audio_input_id_per_second": 25.0,
229
+ "audio_kernel_size": 3,
230
+ "audio_n_mels": 128,
231
+ "audio_nfft": 960,
232
+ "audio_sampling_rate": 24000,
233
+ "audio_segment_size": 6000,
234
+ "audio_start_token_id": 151673,
235
+ "audio_stride_size": 2,
236
+ "audio_token_id": 151669,
237
+ "audio_window_size": 960,
238
+ "audio_zeroemb_idx": [
239
+ 1024,
240
+ 1024,
241
+ 1024,
242
+ 1024,
243
+ 1024,
244
+ 1024,
245
+ 1024,
246
+ 1024,
247
+ 1024,
248
+ 1024,
249
+ 1024,
250
+ 1024,
251
+ 1024,
252
+ 1024,
253
+ 1024,
254
+ 1024,
255
+ 1024,
256
+ 1024,
257
+ 1024,
258
+ 1024
259
+ ],
260
+ "fps": 1.0,
261
+ "image_max_pixels": 8388608,
262
+ "image_min_pixels": 8192,
263
+ "image_token_id": 151655,
264
+ "max_frames": 3600,
265
+ "merge_size": 2,
266
+ "min_frames": null,
267
+ "num_frames": null,
268
+ "pad_token_id": 151643,
269
+ "patch_size": 16,
270
+ "rope_type": "rope",
271
+ "temporal_compression_ratio": 1,
272
+ "temporal_patch_size": 2,
273
+ "use_per_grid_t_timestamps": false,
274
+ "use_video_timestamps": true,
275
+ "video_audio_interleave_length": 0.0,
276
+ "video_end_token_id": 151671,
277
+ "video_max_pixels": 8388608,
278
+ "video_min_pixels": 8192,
279
+ "video_process_num_threads": 16,
280
+ "video_start_token_id": 151670,
281
+ "video_token_id": 151656,
282
+ "video_tokens_per_second": 2,
283
+ "video_total_max_pixels": 268435456,
284
+ "vision_end_token_id": 151653,
285
+ "vision_start_token_id": 151652
286
+ },
287
+ "image_token_id": 151655,
288
+ "video_token_id": 151656,
289
+ "vision_start_token_id": 151652,
290
+ "vision_end_token_id": 151653,
291
+ "vision_model_type": "mimovl",
292
+ "audio_token_id": 151669,
293
+ "audio_start_token_id": 151673,
294
+ "audio_end_token_id": 151674,
295
+ "transformers_version": "5.6.2",
296
+ "architectures": [
297
+ "MiMoV2ForCausalLM"
298
+ ],
299
+ "output_hidden_states": false,
300
+ "return_dict": true,
301
+ "dtype": "bfloat16",
302
+ "chunk_size_feed_forward": 0,
303
+ "is_encoder_decoder": false,
304
+ "id2label": {
305
+ "0": "LABEL_0",
306
+ "1": "LABEL_1"
307
+ },
308
+ "label2id": {
309
+ "LABEL_0": 0,
310
+ "LABEL_1": 1
311
+ },
312
+ "problem_type": null,
313
+ "_name_or_path": "",
314
+ "tie_word_embeddings": false,
315
+ "attention_chunk_size": 128,
316
+ "eos_token_id": 151645,
317
+ "model_type": "mimo_v2",
318
+ "n_shared_experts": null,
319
+ "pad_token_id": 151643,
320
+ "layer_types": [
321
+ "full_attention",
322
+ "sliding_attention",
323
+ "sliding_attention",
324
+ "sliding_attention",
325
+ "sliding_attention",
326
+ "full_attention",
327
+ "sliding_attention",
328
+ "sliding_attention",
329
+ "sliding_attention",
330
+ "sliding_attention",
331
+ "sliding_attention",
332
+ "full_attention",
333
+ "sliding_attention",
334
+ "sliding_attention",
335
+ "sliding_attention",
336
+ "sliding_attention",
337
+ "sliding_attention",
338
+ "full_attention",
339
+ "sliding_attention",
340
+ "sliding_attention",
341
+ "sliding_attention",
342
+ "sliding_attention",
343
+ "sliding_attention",
344
+ "full_attention",
345
+ "sliding_attention",
346
+ "sliding_attention",
347
+ "sliding_attention",
348
+ "sliding_attention",
349
+ "sliding_attention",
350
+ "full_attention",
351
+ "sliding_attention",
352
+ "sliding_attention",
353
+ "sliding_attention",
354
+ "sliding_attention",
355
+ "sliding_attention",
356
+ "full_attention",
357
+ "sliding_attention",
358
+ "sliding_attention",
359
+ "sliding_attention",
360
+ "sliding_attention",
361
+ "sliding_attention",
362
+ "full_attention",
363
+ "sliding_attention",
364
+ "sliding_attention",
365
+ "sliding_attention",
366
+ "sliding_attention",
367
+ "sliding_attention",
368
+ "full_attention"
369
+ ],
370
+ "output_attentions": false,
371
+ "quantization_config": {
372
+ "config_groups": {
373
+ "group_0": {
374
+ "input_activations": {
375
+ "dynamic": false,
376
+ "num_bits": 4,
377
+ "type": "float",
378
+ "group_size": 16
379
+ },
380
+ "weights": {
381
+ "dynamic": false,
382
+ "num_bits": 4,
383
+ "type": "float",
384
+ "group_size": 16
385
+ },
386
+ "targets": [
387
+ "Linear"
388
+ ]
389
+ }
390
+ },
391
+ "ignore": [
392
+ "audio_encoder*",
393
+ "speech_embeddings*",
394
+ "visual*",
395
+ "lm_head",
396
+ "model.embed_tokens",
397
+ "model.norm",
398
+ "model.rotary_emb*",
399
+ "model.swa_rotary_emb*",
400
+ "model.mtp*",
401
+ "model.layers.0*",
402
+ "model.layers.1.self_attn*",
403
+ "model.layers.1.mlp.gate*",
404
+ "model.layers.1.mlp.shared_expert*",
405
+ "model.layers.1.mlp.shared_expert_gate*",
406
+ "model.layers.2.self_attn*",
407
+ "model.layers.2.mlp.gate*",
408
+ "model.layers.2.mlp.shared_expert*",
409
+ "model.layers.2.mlp.shared_expert_gate*",
410
+ "model.layers.3.self_attn*",
411
+ "model.layers.3.mlp.gate*",
412
+ "model.layers.3.mlp.shared_expert*",
413
+ "model.layers.3.mlp.shared_expert_gate*",
414
+ "model.layers.4.self_attn*",
415
+ "model.layers.4.mlp.gate*",
416
+ "model.layers.4.mlp.shared_expert*",
417
+ "model.layers.4.mlp.shared_expert_gate*",
418
+ "model.layers.5.self_attn*",
419
+ "model.layers.5.mlp.gate*",
420
+ "model.layers.5.mlp.shared_expert*",
421
+ "model.layers.5.mlp.shared_expert_gate*",
422
+ "model.layers.6.self_attn*",
423
+ "model.layers.6.mlp.gate*",
424
+ "model.layers.6.mlp.shared_expert*",
425
+ "model.layers.6.mlp.shared_expert_gate*",
426
+ "model.layers.7.self_attn*",
427
+ "model.layers.7.mlp.gate*",
428
+ "model.layers.7.mlp.shared_expert*",
429
+ "model.layers.7.mlp.shared_expert_gate*",
430
+ "model.layers.8.self_attn*",
431
+ "model.layers.8.mlp.gate*",
432
+ "model.layers.8.mlp.shared_expert*",
433
+ "model.layers.8.mlp.shared_expert_gate*",
434
+ "model.layers.9.self_attn*",
435
+ "model.layers.9.mlp.gate*",
436
+ "model.layers.9.mlp.shared_expert*",
437
+ "model.layers.9.mlp.shared_expert_gate*",
438
+ "model.layers.10.self_attn*",
439
+ "model.layers.10.mlp.gate*",
440
+ "model.layers.10.mlp.shared_expert*",
441
+ "model.layers.10.mlp.shared_expert_gate*",
442
+ "model.layers.11.self_attn*",
443
+ "model.layers.11.mlp.gate*",
444
+ "model.layers.11.mlp.shared_expert*",
445
+ "model.layers.11.mlp.shared_expert_gate*",
446
+ "model.layers.12.self_attn*",
447
+ "model.layers.12.mlp.gate*",
448
+ "model.layers.12.mlp.shared_expert*",
449
+ "model.layers.12.mlp.shared_expert_gate*",
450
+ "model.layers.13.self_attn*",
451
+ "model.layers.13.mlp.gate*",
452
+ "model.layers.13.mlp.shared_expert*",
453
+ "model.layers.13.mlp.shared_expert_gate*",
454
+ "model.layers.14.self_attn*",
455
+ "model.layers.14.mlp.gate*",
456
+ "model.layers.14.mlp.shared_expert*",
457
+ "model.layers.14.mlp.shared_expert_gate*",
458
+ "model.layers.15.self_attn*",
459
+ "model.layers.15.mlp.gate*",
460
+ "model.layers.15.mlp.shared_expert*",
461
+ "model.layers.15.mlp.shared_expert_gate*",
462
+ "model.layers.16.self_attn*",
463
+ "model.layers.16.mlp.gate*",
464
+ "model.layers.16.mlp.shared_expert*",
465
+ "model.layers.16.mlp.shared_expert_gate*",
466
+ "model.layers.17.self_attn*",
467
+ "model.layers.17.mlp.gate*",
468
+ "model.layers.17.mlp.shared_expert*",
469
+ "model.layers.17.mlp.shared_expert_gate*",
470
+ "model.layers.18.self_attn*",
471
+ "model.layers.18.mlp.gate*",
472
+ "model.layers.18.mlp.shared_expert*",
473
+ "model.layers.18.mlp.shared_expert_gate*",
474
+ "model.layers.19.self_attn*",
475
+ "model.layers.19.mlp.gate*",
476
+ "model.layers.19.mlp.shared_expert*",
477
+ "model.layers.19.mlp.shared_expert_gate*",
478
+ "model.layers.20.self_attn*",
479
+ "model.layers.20.mlp.gate*",
480
+ "model.layers.20.mlp.shared_expert*",
481
+ "model.layers.20.mlp.shared_expert_gate*",
482
+ "model.layers.21.self_attn*",
483
+ "model.layers.21.mlp.gate*",
484
+ "model.layers.21.mlp.shared_expert*",
485
+ "model.layers.21.mlp.shared_expert_gate*",
486
+ "model.layers.22.self_attn*",
487
+ "model.layers.22.mlp.gate*",
488
+ "model.layers.22.mlp.shared_expert*",
489
+ "model.layers.22.mlp.shared_expert_gate*",
490
+ "model.layers.23.self_attn*",
491
+ "model.layers.23.mlp.gate*",
492
+ "model.layers.23.mlp.shared_expert*",
493
+ "model.layers.23.mlp.shared_expert_gate*",
494
+ "model.layers.24.self_attn*",
495
+ "model.layers.24.mlp.gate*",
496
+ "model.layers.24.mlp.shared_expert*",
497
+ "model.layers.24.mlp.shared_expert_gate*",
498
+ "model.layers.25.self_attn*",
499
+ "model.layers.25.mlp.gate*",
500
+ "model.layers.25.mlp.shared_expert*",
501
+ "model.layers.25.mlp.shared_expert_gate*",
502
+ "model.layers.26.self_attn*",
503
+ "model.layers.26.mlp.gate*",
504
+ "model.layers.26.mlp.shared_expert*",
505
+ "model.layers.26.mlp.shared_expert_gate*",
506
+ "model.layers.27.self_attn*",
507
+ "model.layers.27.mlp.gate*",
508
+ "model.layers.27.mlp.shared_expert*",
509
+ "model.layers.27.mlp.shared_expert_gate*",
510
+ "model.layers.28.self_attn*",
511
+ "model.layers.28.mlp.gate*",
512
+ "model.layers.28.mlp.shared_expert*",
513
+ "model.layers.28.mlp.shared_expert_gate*",
514
+ "model.layers.29.self_attn*",
515
+ "model.layers.29.mlp.gate*",
516
+ "model.layers.29.mlp.shared_expert*",
517
+ "model.layers.29.mlp.shared_expert_gate*",
518
+ "model.layers.30.self_attn*",
519
+ "model.layers.30.mlp.gate*",
520
+ "model.layers.30.mlp.shared_expert*",
521
+ "model.layers.30.mlp.shared_expert_gate*",
522
+ "model.layers.31.self_attn*",
523
+ "model.layers.31.mlp.gate*",
524
+ "model.layers.31.mlp.shared_expert*",
525
+ "model.layers.31.mlp.shared_expert_gate*",
526
+ "model.layers.32.self_attn*",
527
+ "model.layers.32.mlp.gate*",
528
+ "model.layers.32.mlp.shared_expert*",
529
+ "model.layers.32.mlp.shared_expert_gate*",
530
+ "model.layers.33.self_attn*",
531
+ "model.layers.33.mlp.gate*",
532
+ "model.layers.33.mlp.shared_expert*",
533
+ "model.layers.33.mlp.shared_expert_gate*",
534
+ "model.layers.34.self_attn*",
535
+ "model.layers.34.mlp.gate*",
536
+ "model.layers.34.mlp.shared_expert*",
537
+ "model.layers.34.mlp.shared_expert_gate*",
538
+ "model.layers.35.self_attn*",
539
+ "model.layers.35.mlp.gate*",
540
+ "model.layers.35.mlp.shared_expert*",
541
+ "model.layers.35.mlp.shared_expert_gate*",
542
+ "model.layers.36.self_attn*",
543
+ "model.layers.36.mlp.gate*",
544
+ "model.layers.36.mlp.shared_expert*",
545
+ "model.layers.36.mlp.shared_expert_gate*",
546
+ "model.layers.37.self_attn*",
547
+ "model.layers.37.mlp.gate*",
548
+ "model.layers.37.mlp.shared_expert*",
549
+ "model.layers.37.mlp.shared_expert_gate*",
550
+ "model.layers.38.self_attn*",
551
+ "model.layers.38.mlp.gate*",
552
+ "model.layers.38.mlp.shared_expert*",
553
+ "model.layers.38.mlp.shared_expert_gate*",
554
+ "model.layers.39.self_attn*",
555
+ "model.layers.39.mlp.gate*",
556
+ "model.layers.39.mlp.shared_expert*",
557
+ "model.layers.39.mlp.shared_expert_gate*",
558
+ "model.layers.40.self_attn*",
559
+ "model.layers.40.mlp.gate*",
560
+ "model.layers.40.mlp.shared_expert*",
561
+ "model.layers.40.mlp.shared_expert_gate*",
562
+ "model.layers.41.self_attn*",
563
+ "model.layers.41.mlp.gate*",
564
+ "model.layers.41.mlp.shared_expert*",
565
+ "model.layers.41.mlp.shared_expert_gate*",
566
+ "model.layers.42.self_attn*",
567
+ "model.layers.42.mlp.gate*",
568
+ "model.layers.42.mlp.shared_expert*",
569
+ "model.layers.42.mlp.shared_expert_gate*",
570
+ "model.layers.43.self_attn*",
571
+ "model.layers.43.mlp.gate*",
572
+ "model.layers.43.mlp.shared_expert*",
573
+ "model.layers.43.mlp.shared_expert_gate*",
574
+ "model.layers.44.self_attn*",
575
+ "model.layers.44.mlp.gate*",
576
+ "model.layers.44.mlp.shared_expert*",
577
+ "model.layers.44.mlp.shared_expert_gate*",
578
+ "model.layers.45.self_attn*",
579
+ "model.layers.45.mlp.gate*",
580
+ "model.layers.45.mlp.shared_expert*",
581
+ "model.layers.45.mlp.shared_expert_gate*",
582
+ "model.layers.46.self_attn*",
583
+ "model.layers.46.mlp.gate*",
584
+ "model.layers.46.mlp.shared_expert*",
585
+ "model.layers.46.mlp.shared_expert_gate*",
586
+ "model.layers.47.self_attn*",
587
+ "model.layers.47.mlp.gate*",
588
+ "model.layers.47.mlp.shared_expert*",
589
+ "model.layers.47.mlp.shared_expert_gate*"
590
+ ],
591
+ "quant_algo": "NVFP4",
592
+ "producer": {
593
+ "name": "modelopt",
594
+ "version": "0.39.0.dev290+gf9d9a71de.d20260407"
595
+ },
596
+ "quant_method": "modelopt"
597
+ }
598
+ }
generation_config.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token_id": 151643,
3
+ "do_sample": false,
4
+ "eos_token_id": [151643, 151645, 1561672],
5
+ "temperature": 1.0,
6
+ "top_p": 0.95,
7
+ "max_new_tokens": 2048,
8
+ "transformers_version": "4.37.0"
9
+ }
hf_quant_config.json ADDED
@@ -0,0 +1,227 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "config_groups": {
3
+ "group_0": {
4
+ "input_activations": {
5
+ "dynamic": false,
6
+ "num_bits": 4,
7
+ "type": "float",
8
+ "group_size": 16
9
+ },
10
+ "weights": {
11
+ "dynamic": false,
12
+ "num_bits": 4,
13
+ "type": "float",
14
+ "group_size": 16
15
+ },
16
+ "targets": [
17
+ "Linear"
18
+ ]
19
+ }
20
+ },
21
+ "ignore": [
22
+ "audio_encoder*",
23
+ "speech_embeddings*",
24
+ "visual*",
25
+ "lm_head",
26
+ "model.embed_tokens",
27
+ "model.norm",
28
+ "model.rotary_emb*",
29
+ "model.swa_rotary_emb*",
30
+ "model.mtp*",
31
+ "model.layers.0*",
32
+ "model.layers.1.self_attn*",
33
+ "model.layers.1.mlp.gate*",
34
+ "model.layers.1.mlp.shared_expert*",
35
+ "model.layers.1.mlp.shared_expert_gate*",
36
+ "model.layers.2.self_attn*",
37
+ "model.layers.2.mlp.gate*",
38
+ "model.layers.2.mlp.shared_expert*",
39
+ "model.layers.2.mlp.shared_expert_gate*",
40
+ "model.layers.3.self_attn*",
41
+ "model.layers.3.mlp.gate*",
42
+ "model.layers.3.mlp.shared_expert*",
43
+ "model.layers.3.mlp.shared_expert_gate*",
44
+ "model.layers.4.self_attn*",
45
+ "model.layers.4.mlp.gate*",
46
+ "model.layers.4.mlp.shared_expert*",
47
+ "model.layers.4.mlp.shared_expert_gate*",
48
+ "model.layers.5.self_attn*",
49
+ "model.layers.5.mlp.gate*",
50
+ "model.layers.5.mlp.shared_expert*",
51
+ "model.layers.5.mlp.shared_expert_gate*",
52
+ "model.layers.6.self_attn*",
53
+ "model.layers.6.mlp.gate*",
54
+ "model.layers.6.mlp.shared_expert*",
55
+ "model.layers.6.mlp.shared_expert_gate*",
56
+ "model.layers.7.self_attn*",
57
+ "model.layers.7.mlp.gate*",
58
+ "model.layers.7.mlp.shared_expert*",
59
+ "model.layers.7.mlp.shared_expert_gate*",
60
+ "model.layers.8.self_attn*",
61
+ "model.layers.8.mlp.gate*",
62
+ "model.layers.8.mlp.shared_expert*",
63
+ "model.layers.8.mlp.shared_expert_gate*",
64
+ "model.layers.9.self_attn*",
65
+ "model.layers.9.mlp.gate*",
66
+ "model.layers.9.mlp.shared_expert*",
67
+ "model.layers.9.mlp.shared_expert_gate*",
68
+ "model.layers.10.self_attn*",
69
+ "model.layers.10.mlp.gate*",
70
+ "model.layers.10.mlp.shared_expert*",
71
+ "model.layers.10.mlp.shared_expert_gate*",
72
+ "model.layers.11.self_attn*",
73
+ "model.layers.11.mlp.gate*",
74
+ "model.layers.11.mlp.shared_expert*",
75
+ "model.layers.11.mlp.shared_expert_gate*",
76
+ "model.layers.12.self_attn*",
77
+ "model.layers.12.mlp.gate*",
78
+ "model.layers.12.mlp.shared_expert*",
79
+ "model.layers.12.mlp.shared_expert_gate*",
80
+ "model.layers.13.self_attn*",
81
+ "model.layers.13.mlp.gate*",
82
+ "model.layers.13.mlp.shared_expert*",
83
+ "model.layers.13.mlp.shared_expert_gate*",
84
+ "model.layers.14.self_attn*",
85
+ "model.layers.14.mlp.gate*",
86
+ "model.layers.14.mlp.shared_expert*",
87
+ "model.layers.14.mlp.shared_expert_gate*",
88
+ "model.layers.15.self_attn*",
89
+ "model.layers.15.mlp.gate*",
90
+ "model.layers.15.mlp.shared_expert*",
91
+ "model.layers.15.mlp.shared_expert_gate*",
92
+ "model.layers.16.self_attn*",
93
+ "model.layers.16.mlp.gate*",
94
+ "model.layers.16.mlp.shared_expert*",
95
+ "model.layers.16.mlp.shared_expert_gate*",
96
+ "model.layers.17.self_attn*",
97
+ "model.layers.17.mlp.gate*",
98
+ "model.layers.17.mlp.shared_expert*",
99
+ "model.layers.17.mlp.shared_expert_gate*",
100
+ "model.layers.18.self_attn*",
101
+ "model.layers.18.mlp.gate*",
102
+ "model.layers.18.mlp.shared_expert*",
103
+ "model.layers.18.mlp.shared_expert_gate*",
104
+ "model.layers.19.self_attn*",
105
+ "model.layers.19.mlp.gate*",
106
+ "model.layers.19.mlp.shared_expert*",
107
+ "model.layers.19.mlp.shared_expert_gate*",
108
+ "model.layers.20.self_attn*",
109
+ "model.layers.20.mlp.gate*",
110
+ "model.layers.20.mlp.shared_expert*",
111
+ "model.layers.20.mlp.shared_expert_gate*",
112
+ "model.layers.21.self_attn*",
113
+ "model.layers.21.mlp.gate*",
114
+ "model.layers.21.mlp.shared_expert*",
115
+ "model.layers.21.mlp.shared_expert_gate*",
116
+ "model.layers.22.self_attn*",
117
+ "model.layers.22.mlp.gate*",
118
+ "model.layers.22.mlp.shared_expert*",
119
+ "model.layers.22.mlp.shared_expert_gate*",
120
+ "model.layers.23.self_attn*",
121
+ "model.layers.23.mlp.gate*",
122
+ "model.layers.23.mlp.shared_expert*",
123
+ "model.layers.23.mlp.shared_expert_gate*",
124
+ "model.layers.24.self_attn*",
125
+ "model.layers.24.mlp.gate*",
126
+ "model.layers.24.mlp.shared_expert*",
127
+ "model.layers.24.mlp.shared_expert_gate*",
128
+ "model.layers.25.self_attn*",
129
+ "model.layers.25.mlp.gate*",
130
+ "model.layers.25.mlp.shared_expert*",
131
+ "model.layers.25.mlp.shared_expert_gate*",
132
+ "model.layers.26.self_attn*",
133
+ "model.layers.26.mlp.gate*",
134
+ "model.layers.26.mlp.shared_expert*",
135
+ "model.layers.26.mlp.shared_expert_gate*",
136
+ "model.layers.27.self_attn*",
137
+ "model.layers.27.mlp.gate*",
138
+ "model.layers.27.mlp.shared_expert*",
139
+ "model.layers.27.mlp.shared_expert_gate*",
140
+ "model.layers.28.self_attn*",
141
+ "model.layers.28.mlp.gate*",
142
+ "model.layers.28.mlp.shared_expert*",
143
+ "model.layers.28.mlp.shared_expert_gate*",
144
+ "model.layers.29.self_attn*",
145
+ "model.layers.29.mlp.gate*",
146
+ "model.layers.29.mlp.shared_expert*",
147
+ "model.layers.29.mlp.shared_expert_gate*",
148
+ "model.layers.30.self_attn*",
149
+ "model.layers.30.mlp.gate*",
150
+ "model.layers.30.mlp.shared_expert*",
151
+ "model.layers.30.mlp.shared_expert_gate*",
152
+ "model.layers.31.self_attn*",
153
+ "model.layers.31.mlp.gate*",
154
+ "model.layers.31.mlp.shared_expert*",
155
+ "model.layers.31.mlp.shared_expert_gate*",
156
+ "model.layers.32.self_attn*",
157
+ "model.layers.32.mlp.gate*",
158
+ "model.layers.32.mlp.shared_expert*",
159
+ "model.layers.32.mlp.shared_expert_gate*",
160
+ "model.layers.33.self_attn*",
161
+ "model.layers.33.mlp.gate*",
162
+ "model.layers.33.mlp.shared_expert*",
163
+ "model.layers.33.mlp.shared_expert_gate*",
164
+ "model.layers.34.self_attn*",
165
+ "model.layers.34.mlp.gate*",
166
+ "model.layers.34.mlp.shared_expert*",
167
+ "model.layers.34.mlp.shared_expert_gate*",
168
+ "model.layers.35.self_attn*",
169
+ "model.layers.35.mlp.gate*",
170
+ "model.layers.35.mlp.shared_expert*",
171
+ "model.layers.35.mlp.shared_expert_gate*",
172
+ "model.layers.36.self_attn*",
173
+ "model.layers.36.mlp.gate*",
174
+ "model.layers.36.mlp.shared_expert*",
175
+ "model.layers.36.mlp.shared_expert_gate*",
176
+ "model.layers.37.self_attn*",
177
+ "model.layers.37.mlp.gate*",
178
+ "model.layers.37.mlp.shared_expert*",
179
+ "model.layers.37.mlp.shared_expert_gate*",
180
+ "model.layers.38.self_attn*",
181
+ "model.layers.38.mlp.gate*",
182
+ "model.layers.38.mlp.shared_expert*",
183
+ "model.layers.38.mlp.shared_expert_gate*",
184
+ "model.layers.39.self_attn*",
185
+ "model.layers.39.mlp.gate*",
186
+ "model.layers.39.mlp.shared_expert*",
187
+ "model.layers.39.mlp.shared_expert_gate*",
188
+ "model.layers.40.self_attn*",
189
+ "model.layers.40.mlp.gate*",
190
+ "model.layers.40.mlp.shared_expert*",
191
+ "model.layers.40.mlp.shared_expert_gate*",
192
+ "model.layers.41.self_attn*",
193
+ "model.layers.41.mlp.gate*",
194
+ "model.layers.41.mlp.shared_expert*",
195
+ "model.layers.41.mlp.shared_expert_gate*",
196
+ "model.layers.42.self_attn*",
197
+ "model.layers.42.mlp.gate*",
198
+ "model.layers.42.mlp.shared_expert*",
199
+ "model.layers.42.mlp.shared_expert_gate*",
200
+ "model.layers.43.self_attn*",
201
+ "model.layers.43.mlp.gate*",
202
+ "model.layers.43.mlp.shared_expert*",
203
+ "model.layers.43.mlp.shared_expert_gate*",
204
+ "model.layers.44.self_attn*",
205
+ "model.layers.44.mlp.gate*",
206
+ "model.layers.44.mlp.shared_expert*",
207
+ "model.layers.44.mlp.shared_expert_gate*",
208
+ "model.layers.45.self_attn*",
209
+ "model.layers.45.mlp.gate*",
210
+ "model.layers.45.mlp.shared_expert*",
211
+ "model.layers.45.mlp.shared_expert_gate*",
212
+ "model.layers.46.self_attn*",
213
+ "model.layers.46.mlp.gate*",
214
+ "model.layers.46.mlp.shared_expert*",
215
+ "model.layers.46.mlp.shared_expert_gate*",
216
+ "model.layers.47.self_attn*",
217
+ "model.layers.47.mlp.gate*",
218
+ "model.layers.47.mlp.shared_expert*",
219
+ "model.layers.47.mlp.shared_expert_gate*"
220
+ ],
221
+ "quant_algo": "NVFP4",
222
+ "producer": {
223
+ "name": "modelopt",
224
+ "version": "0.39.0.dev290+gf9d9a71de.d20260407"
225
+ },
226
+ "quant_method": "modelopt"
227
+ }
merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
model-00001-of-00035.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:42f4a5e9de1c0cd8995f6d75e8bb42bbf196f7fe3daf513d5fd75e9d91bd24f3
3
+ size 5372235616
model-00002-of-00035.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:78d1330821c90e43611ca6e1dce53518576e2025ed4191e64ade4aeb1bff2b56
3
+ size 5372276856
model-00003-of-00035.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6bc20dc3244d480df62d5ab262bbb1309c9b88b79871e271775795458393edfd
3
+ size 5369134748
model-00004-of-00035.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:86357f39be6d496eea2dba6993968bdf88ed3492d901988d6d625c6794a28022
3
+ size 5371751960
model-00005-of-00035.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ba6e6048e673c988c7c7c1257c5e347e17a3a5e29bb7623e27151dc9eed1f8e2
3
+ size 5372277256
model-00006-of-00035.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b5bb51fcfce17ba122676a11fa1808e8f2ed9f5f380714a5fbd1d93bba313589
3
+ size 5369657836
model-00007-of-00035.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b00eff0e9b5227e562fc0304a366461e18bef2a8925d6f8e1029ccc0b1bd7134
3
+ size 5372278568
model-00008-of-00035.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:38f556021f59a2be6420b8e62f1f5cd54826cdc2e27f6cb64d6e1a1456f2cfb3
3
+ size 5369137812
model-00009-of-00035.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:458f6df1c89d86338d78f9c779805300ab541a86b2dcf835a39600cff9299b57
3
+ size 5371755392
model-00010-of-00035.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0fdb0484cf7ac98c4b3908d02b7c38d222b4696893938bd556fbc75fd806fb89
3
+ size 5372280552
model-00011-of-00035.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:45237e6f1bff7e92e93246c4eea4fb599a667f1dd2199e85be42f753c1e3c3a6
3
+ size 5369660924
model-00012-of-00035.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2d5f00ed0345198d25164aa0f1c03eea5d0e463b6d644a4e76f304a692ee9dea
3
+ size 5371232280
model-00013-of-00035.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1b13d6d07df72ad3d3d9789fe188d8ca68862910496154802bdd698db66a3ba1
3
+ size 5369661364
model-00014-of-00035.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:37bc51697a9c91a195a1a2aceeca3a7e225900fd4ad2bfd62468de60e110243a
3
+ size 5372279920
model-00015-of-00035.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2f4255f3df87060f6d34b2b0ec57e52380fb4354c26c6b582a1f969dae234cdb
3
+ size 5372280552
model-00016-of-00035.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dc79fd5e3de1900f4e457acb74c7f992236bbe8eca262e217f97b210899fd187
3
+ size 5369137156
model-00017-of-00035.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b1d0b014414b735cdd56f593562f3e5131f1131d842eb16ff84cc2ca0a004082
3
+ size 5371756048
model-00018-of-00035.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:89b994431fa99e7292f3fa9aca1f3b2365b222da65ddd1b9c0331ee84d370816
3
+ size 5369661260
model-00019-of-00035.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:457db02428d00fc2c830dd12c8fd311d2eb9572b7f08d15c1a3e461a0a18e239
3
+ size 5372280016
model-00020-of-00035.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:94702b768575ea7a72149c072380d75e9820c41ba3f7e5593a421e667db518c7
3
+ size 5420003292
model-00021-of-00035.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f4efab9985e73460ab0fdfc9cb495c47d5b0ed347a60df15985050bcf7af45aa
3
+ size 5372798532
model-00022-of-00035.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0ae2540f44f8034d9bb90aaa7f9c647b38f58e94cb73fb01d50525704a12ff65
3
+ size 5372280552
model-00023-of-00035.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f6631c938bddc3af941725d838b0a541dcd7af0fdfdb6621abc5ef1c1d8f8a79
3
+ size 5369661124
model-00024-of-00035.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:684f0ae022b9a080066925be1a080a75a8c46198250066275ac73b4622e486a9
3
+ size 5372280152
model-00025-of-00035.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8a47ac8e277d517378e5d06c1be4d9a5d44249ced7dfed42bc77c475e761e146
3
+ size 5369137940
model-00026-of-00035.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:56bfe09813ab025e863b437d1be054a8375724f708db0d9aa3c7a0c25312237c
3
+ size 5371755264
model-00027-of-00035.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a85d5d7a6fbbbe77c34a9e96f36423a39601df27053cbbad297431e3ca518c29
3
+ size 5372280552
model-00028-of-00035.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ccf4f830a8d1f47b08398aceeeb23a691a787b7bd34f5f08eb832f603404310b
3
+ size 5369661020
model-00029-of-00035.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bf2bc24ad22ec8178e8d19b6fb85953e6c458efcf79c1dca43a83c38ee57edf0
3
+ size 5371232184
model-00030-of-00035.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fa1cac4a8e77f175655461743e6aba093b41de5341ecd852afa2e62a2722c4ce
3
+ size 5369661460
model-00031-of-00035.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5d2239ec3e46327ff2fd995c25856a70efa83049c4560e6616b1d3d1cf76fc64
3
+ size 5372279816
model-00032-of-00035.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a04035695604ad201b8885a23153af81d3ccbc3bcfa8bc27c17bb5959c8007ad
3
+ size 5372280552
model-00033-of-00035.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ff8ae119fec50f327828bac6e6c3bb93eb91e6f2b00f0f64a01dd5f60d2d5d67
3
+ size 5369137260
model-00034-of-00035.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e39b511533abf5bc8f4398096ded7336797aeeda642596aeaf991a05f8376d3b
3
+ size 5376279968
model-00035-of-00035.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:35bc1a3dc07112ece7537475cc4d2626aed53de23fe1d4d6594a74668e172818
3
+ size 1922097992
model-inputscales.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:381f70f31092063586d99f91a7a5ed82f8974116050e8276e43a8cb3782199d6
3
+ size 4157304
model-mtp.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c6c24e326f2bdba86afc2b2667787d400fc956e81a8238cf6a8a82334747f009
3
+ size 1975644568
model.safetensors.index.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:55b19ae8ec08d5369ecfbae408d22d0607bebee688fa78defd69b47a37c986c0
3
+ size 13743417