Add files using upload-large-folder tool

README.md

@@ -0,0 +1,48 @@
+---
+language:
+- en
+- fr
+- de
+- es
+- it
+- pt
+- zh
+- ja
+- ru
+- ko
+license: other
+license_name: mrl
+base_model: mistralai/Pixtral-Large-Instruct-2411
+base_model_relation: quantized
+inference: false
+license_link: https://mistral.ai/licenses/MRL-0.1.md
+library_name: transformers
+pipeline_tag: image-text-to-text
+---
+
+# Pixtral-Large-Instruct-2411 🧡 ExLlamaV2 2.5bpw Quant
+
+2.5bpw quant of [Pixtral-Large-Instruct](https://huggingface.co/nintwentydo/Pixtral-Large-Instruct-2411).  
+
+Vision inputs working on dev branch of [ExLlamaV2](https://github.com/turboderp/exllamav2/tree/dev).  
+
+
+## Tokenizer And Prompt Template
+Using conversion of v7m1 tokenizer with 32k vocab size.  
+ 
+Chat template in chat_template.json uses the v7 instruct template:
+ 
+```
+<s>[SYSTEM_PROMPT] <system prompt>[/SYSTEM_PROMPT][INST] <user message>[/INST] <assistant response></s>[INST] <user message>[/INST]
+```
+
+## Available Sizes
+| Repo | Bits | Head Bits | Size |
+| ----------- | ------ | ------ | ------ |
+| [nintwentydo/Pixtral-Large-Instruct-2411-exl2-2.0bpw](https://huggingface.co/nintwentydo/Pixtral-Large-Instruct-2411-exl2-2.0bpw) | 2.0 | 6.0 | 35.18 GB |
+| [nintwentydo/Pixtral-Large-Instruct-2411-exl2-2.5bpw](https://huggingface.co/nintwentydo/Pixtral-Large-Instruct-2411-exl2-2.5bpw) | 2.5 | 6.0 | 39.34 GB |
+| [nintwentydo/Pixtral-Large-Instruct-2411-exl2-3.0bpw](https://huggingface.co/nintwentydo/Pixtral-Large-Instruct-2411-exl2-3.0bpw) | 3.0 | 6.0 | 46.42 GB |
+| [nintwentydo/Pixtral-Large-Instruct-2411-exl2-3.5bpw](https://huggingface.co/nintwentydo/Pixtral-Large-Instruct-2411-exl2-3.5bpw) | 3.5 | 6.0 | 53.50 GB |
+| [nintwentydo/Pixtral-Large-Instruct-2411-exl2-4.0bpw](https://huggingface.co/nintwentydo/Pixtral-Large-Instruct-2411-exl2-4.0bpw) | 4.0 | 6.0 | 60.61GB |
+| [nintwentydo/Pixtral-Large-Instruct-2411-exl2-5.0bpw](https://huggingface.co/nintwentydo/Pixtral-Large-Instruct-2411-exl2-5.0bpw) | 5.0 | 6.0 | 74.76 GB |
+| [nintwentydo/Pixtral-Large-Instruct-2411-exl2-6.0bpw](https://huggingface.co/nintwentydo/Pixtral-Large-Instruct-2411-exl2-6.0bpw) | 6.0 | 8.0 | 88.81GB |

chat_template.json

@@ -0,0 +1,3 @@
+{
+    "chat_template": "{{- bos_token }}    \n{%- for message in messages %}    \n    {%- if message['role'] == 'user' %}    \n        {{- '[INST]' + ' ' }}    \n        {%- if message['content'] is not string %}    \n            {%- for chunk in message['content'] %}    \n                {%- if chunk['type'] == 'text' %}    \n                    {{- chunk['content'] }}    \n                {%- elif chunk['type'] == 'image' %}    \n                    {{- '[IMG]' }}    \n                {%- else %}    \n                    {{- raise_exception('Unrecognized content type!') }}    \n                {%- endif %}    \n            {%- endfor %}    \n                {%- else %}    \n                    {{- message['content'] }}    \n        {%- endif %}    \n            {{- '[\/INST]' }}    \n    {%- if not loop.last and messages[loop.index]['role'] == 'user' %}    \n        {{- eos_token }}    \n    {%- endif %}    \n    {%- elif message['role'] == 'system' %}    \n        {{- '[SYSTEM_PROMPT] ' + message['content'] + '[\/SYSTEM_PROMPT]' }}    \n    {%- elif message['role'] == 'assistant' %}    \n        {{- ' ' + message['content'] + eos_token }}    \n    {%- else %}    \n        {{- raise_exception('Only user, system and assistant roles are supported!') }}    \n    {%- endif %}    \n{%- endfor %}"
+}

config.json

@@ -0,0 +1,54 @@
+{
+    "architectures": [
+        "LlavaForConditionalGeneration"
+    ],
+    "ignore_index": -100,
+    "image_seq_length": 1,
+    "image_token_index": 10,
+    "model_type": "llava",
+    "multimodal_projector_bias": false,
+    "projector_hidden_act": "gelu",
+    "text_config": {
+        "hidden_size": 12288,
+        "intermediate_size": 28672,
+        "is_composition": true,
+        "max_position_embeddings": 131072,
+        "model_type": "mistral",
+        "norm_eps": 1e-05,
+        "rms_norm_eps": 1e-05,
+        "num_attention_heads": 96,
+        "num_hidden_layers": 88,
+        "num_key_value_heads": 8,
+        "rope_theta": 1000000000.0,
+        "sliding_window": null,
+        "vocab_size": 32768
+    },
+    "torch_dtype": "bfloat16",
+    "transformers_version": "4.47.0.dev0",
+    "vision_config": {
+        "head_dim": 88,
+        "hidden_act": "silu",
+        "hidden_size": 1408,
+        "image_size": 1024,
+        "image_token_id": 10,
+        "intermediate_size": 6144,
+        "model_type": "pixtral",
+        "num_hidden_layers": 40,
+        "num_attention_heads": 16,
+        "patch_size": 16,
+        "rope_theta": 10000.0
+    },
+    "vision_feature_layer": -1,
+    "vision_feature_select_strategy": "full",
+    "quantization_config": {
+        "quant_method": "exl2",
+        "version": "0.2.6",
+        "bits": 3.5,
+        "head_bits": 6,
+        "calibration": {
+            "rows": 115,
+            "length": 2048,
+            "dataset": "(default)"
+        }
+    }
+}

generation_config.json

@@ -0,0 +1,6 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "transformers_version": "4.47.0.dev0"
+}

output-00001-of-00006.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:25761b53966c1a21617ab391849a0852e9c71bd8c563accb2e9c52d8876aef80
+size 10717974532

output-00002-of-00006.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0d642cf0006cde9d743159d2a65eff4419acb0456473851c7c6be5ffade6a9bc
+size 10723013376

output-00003-of-00006.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:04fcb6d2f5495bf9b2b4b55af26f450ad8c548fae9bc05354c394370ac92afcb
+size 10682895892

output-00004-of-00006.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9c3275f6a6bdd3361814f272f34ff1fcdc9593096bc258f74171660ace44e91f
+size 10708460324

output-00005-of-00006.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7dcbdca2d3ca5b4d74ff63a203248b76f70c5cc3d66b9bb157092f5e8ccf13dc
+size 10702098960

output-00006-of-00006.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f4128696802d8cb887b91a5b7e1d7295c14954663517d3527decba454811ea59
+size 3906298528

preprocessor_config.json

@@ -0,0 +1,27 @@
+{
+  "do_convert_rgb": true,
+  "do_normalize": true,
+  "do_rescale": true,
+  "do_resize": true,
+  "image_mean": [
+    0.48145466,
+    0.4578275,
+    0.40821073
+  ],
+  "image_processor_type": "PixtralImageProcessor",
+  "image_std": [
+    0.26862954,
+    0.26130258,
+    0.27577711
+  ],
+  "patch_size": {
+    "height": 16,
+    "width": 16
+  },
+  "processor_class": "PixtralProcessor",
+  "resample": 3,
+  "rescale_factor": 0.00392156862745098,
+  "size": {
+    "longest_edge": 1024
+  }
+}

processor_config.json

@@ -0,0 +1,7 @@
+{
+  "image_break_token": "[IMG_BREAK]",
+  "image_end_token": "[IMG_END]",
+  "image_token": "[IMG]",
+  "patch_size": 16,
+  "processor_class": "PixtralProcessor"
+}

tokenizer.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1b968b8dc352f42192367337c78ccc61e1eaddc6d641a579372d4f20694beb7a
+size 587562