Upload WarMachine 4B weights (qwen3.5-4b)

4B/README.md

@@ -8,13 +8,15 @@ tags:
 - kali-linux
 - security
 - axolotl
+- multimodal
+- vision
 library_name: transformers
-pipeline_tag: text-generation
+pipeline_tag: image-text-to-text
 ---
 
 # WarMachine — 4B Tool-Calling Model
 
-Fine-tuned for **automated penetration testing** using Kali Linux CLI tools.
+Fine-tuned for **automated penetration testing** using Kali Linux CLI tools while preserving the Qwen3.5 multimodal wrapper and processor.
 
 ## What This Model Does
 
@@ -23,23 +25,24 @@ WarMachine is trained to act as a pentesting automation agent that:
 2. **Selects** appropriate tools from the Kali Linux arsenal
 3. **Executes** CLI commands via structured tool calls
 4. **Parses** output to extract actionable intelligence
-5. **Chains** tools together toward an objective
+5. **Can inspect visual inputs** such as screenshots through the base Qwen3.5 vision stack
+6. **Chains** tools together toward an objective
 
 ## Usage
 
 ```python
-from transformers import AutoModelForCausalLM, AutoTokenizer
+from transformers import AutoProcessor, Qwen3_5ForConditionalGeneration
 
-model = AutoModelForCausalLM.from_pretrained(
+processor = AutoProcessor.from_pretrained(
     "ovedrive/WarMachine",
     subfolder="4B",
-    torch_dtype="auto",
-    device_map="auto",
     trust_remote_code=True,
 )
-tokenizer = AutoTokenizer.from_pretrained(
+model = Qwen3_5ForConditionalGeneration.from_pretrained(
     "ovedrive/WarMachine",
     subfolder="4B",
+    torch_dtype="auto",
+    device_map="auto",
     trust_remote_code=True,
 )
 
@@ -47,12 +50,14 @@ messages = [
     {"role": "system", "content": "You are WarMachine, a penetration testing automation agent..."},
     {"role": "user", "content": "Enumerate the target 10.10.10.5"},
 ]
-text = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
-inputs = tokenizer(text, return_tensors="pt").to(model.device)
+text = processor.tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
+inputs = processor(text=text, return_tensors="pt").to(model.device)
 outputs = model.generate(**inputs, max_new_tokens=512)
-print(tokenizer.decode(outputs[0], skip_special_tokens=True))
+print(processor.tokenizer.decode(outputs[0], skip_special_tokens=True))
 ```
 
+If you want the model to inspect a screenshot at inference time, pass the image through the processor as an image input instead of converting the screenshot into plain text yourself.
+
 ## Training Details
 
 - **Base Model:** Qwen/Qwen3.5-4B
@@ -60,12 +65,7 @@ print(tokenizer.decode(outputs[0], skip_special_tokens=True))
 - **Config:** `qwen3.5-4b`
 - **Data:** Synthetic + curated pentesting tool-calling conversations
 - **Focus:** CEH / PNPT / HTB style engagements
-
-## Disclaimer
-
-This model is intended for **authorized security testing and educational purposes only**.
-Always obtain proper authorization before testing systems you do not own.
-
+- **Wrapper Preservation:** trained and merged through the Qwen3.5 multimodal wrapper so the base vision stack remains available
 
 ## Attribution
 
@@ -83,4 +83,10 @@ If you use this model, please cite:
   year         = {2026},
   howpublished = {\url{https://huggingface.co/ovedrive/WarMachine/tree/main/4B}},
   note         = {Author contact: ab@3computing.com}
-}
+}
+```
+
+## Disclaimer
+
+This model is intended for **authorized security testing and educational purposes only**.
+Always obtain proper authorization before testing systems you do not own.

4B/config.json

@@ -1,83 +1,109 @@
 {
   "architectures": [
-    "Qwen3_5ForCausalLM"
+    "Qwen3_5ForConditionalGeneration"
   ],
-  "attention_bias": false,
-  "attention_dropout": 0.0,
-  "attn_output_gate": true,
-  "bos_token_id": null,
   "dtype": "bfloat16",
-  "eos_token_id": 248044,
-  "full_attention_interval": 4,
-  "head_dim": 256,
-  "hidden_act": "silu",
-  "hidden_size": 2560,
-  "initializer_range": 0.02,
-  "intermediate_size": 9216,
-  "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_conv_kernel_dim": 4,
-  "linear_key_head_dim": 128,
-  "linear_num_key_heads": 16,
-  "linear_num_value_heads": 32,
-  "linear_value_head_dim": 128,
-  "mamba_ssm_dtype": "float32",
-  "max_position_embeddings": 262144,
-  "mlp_only_layers": [],
-  "model_type": "qwen3_5_text",
-  "mtp_num_hidden_layers": 1,
-  "mtp_use_dedicated_embeddings": false,
-  "num_attention_heads": 16,
-  "num_hidden_layers": 32,
-  "num_key_value_heads": 4,
-  "pad_token_id": null,
-  "partial_rotary_factor": 0.25,
-  "rms_norm_eps": 1e-06,
-  "rope_parameters": {
-    "mrope_interleaved": true,
-    "mrope_section": [
-      11,
-      11,
-      10
+  "image_token_id": 248056,
+  "model_type": "qwen3_5",
+  "text_config": {
+    "attention_bias": false,
+    "attention_dropout": 0.0,
+    "attn_output_gate": true,
+    "bos_token_id": null,
+    "dtype": "bfloat16",
+    "eos_token_id": 248044,
+    "full_attention_interval": 4,
+    "head_dim": 256,
+    "hidden_act": "silu",
+    "hidden_size": 2560,
+    "initializer_range": 0.02,
+    "intermediate_size": 9216,
+    "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_conv_kernel_dim": 4,
+    "linear_key_head_dim": 128,
+    "linear_num_key_heads": 16,
+    "linear_num_value_heads": 32,
+    "linear_value_head_dim": 128,
+    "mamba_ssm_dtype": "float32",
+    "max_position_embeddings": 262144,
+    "mlp_only_layers": [],
+    "model_type": "qwen3_5_text",
+    "mtp_num_hidden_layers": 1,
+    "mtp_use_dedicated_embeddings": false,
+    "num_attention_heads": 16,
+    "num_hidden_layers": 32,
+    "num_key_value_heads": 4,
+    "pad_token_id": null,
     "partial_rotary_factor": 0.25,
-    "rope_theta": 10000000,
-    "rope_type": "default"
+    "rms_norm_eps": 1e-06,
+    "rope_parameters": {
+      "mrope_interleaved": true,
+      "mrope_section": [
+        11,
+        11,
+        10
+      ],
+      "partial_rotary_factor": 0.25,
+      "rope_theta": 10000000,
+      "rope_type": "default"
+    },
+    "tie_word_embeddings": true,
+    "use_cache": true,
+    "vocab_size": 248320
   },
   "tie_word_embeddings": true,
   "transformers_version": "5.3.0",
-  "use_cache": true,
-  "vocab_size": 248320
+  "video_token_id": 248057,
+  "vision_config": {
+    "deepstack_visual_indexes": [],
+    "depth": 24,
+    "dtype": "bfloat16",
+    "hidden_act": "gelu_pytorch_tanh",
+    "hidden_size": 1024,
+    "in_channels": 3,
+    "initializer_range": 0.02,
+    "intermediate_size": 4096,
+    "model_type": "qwen3_5",
+    "num_heads": 16,
+    "num_position_embeddings": 2304,
+    "out_hidden_size": 2560,
+    "patch_size": 16,
+    "spatial_merge_size": 2,
+    "temporal_patch_size": 2
+  },
+  "vision_end_token_id": 248054,
+  "vision_start_token_id": 248053
 }

4B/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a6d4236e793dee0377971c8c9e2cb656fd17b28e8539ee4b1a621012c82f3d83
-size 8411558400
+oid sha256:82daca3563b75e7219ed03ccce86de2b19f3523e4c1f2311ad26018c535d3f20
+size 9078620536

4B/processor_config.json

@@ -0,0 +1,63 @@
+{
+  "image_processor": {
+    "data_format": "channels_first",
+    "do_convert_rgb": true,
+    "do_normalize": true,
+    "do_rescale": true,
+    "do_resize": true,
+    "image_mean": [
+      0.5,
+      0.5,
+      0.5
+    ],
+    "image_processor_type": "Qwen2VLImageProcessorFast",
+    "image_std": [
+      0.5,
+      0.5,
+      0.5
+    ],
+    "merge_size": 2,
+    "patch_size": 16,
+    "resample": 3,
+    "rescale_factor": 0.00392156862745098,
+    "size": {
+      "longest_edge": 16777216,
+      "shortest_edge": 65536
+    },
+    "temporal_patch_size": 2
+  },
+  "processor_class": "Qwen3VLProcessor",
+  "video_processor": {
+    "data_format": "channels_first",
+    "default_to_square": true,
+    "do_convert_rgb": true,
+    "do_normalize": true,
+    "do_rescale": true,
+    "do_resize": true,
+    "do_sample_frames": true,
+    "fps": 2,
+    "image_mean": [
+      0.5,
+      0.5,
+      0.5
+    ],
+    "image_std": [
+      0.5,
+      0.5,
+      0.5
+    ],
+    "max_frames": 768,
+    "merge_size": 2,
+    "min_frames": 4,
+    "patch_size": 16,
+    "resample": 3,
+    "rescale_factor": 0.00392156862745098,
+    "return_metadata": false,
+    "size": {
+      "longest_edge": 25165824,
+      "shortest_edge": 4096
+    },
+    "temporal_patch_size": 2,
+    "video_processor_type": "Qwen3VLVideoProcessor"
+  }
+}

4B/tokenizer_config.json

@@ -22,6 +22,7 @@
   },
   "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",
   "unk_token": null,