Initial upload of Qwen3-Reranker-4B-AWQ-INT4

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* 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 +1,53 @@
+---
+license: apache-2.0
+
+base_model: Qwen/Qwen3-Reranker-4B
+tags:
+  - quantized
+  - 4-bit
+  - int4
+  - awq
+language:
+  - en
+library_name: transformers
+pipeline_tag: text-generation
+---
+
+# Qwen3-Reranker-4B-AWQ-INT4
+
+INT4 weight-only quantization of [`Qwen/Qwen3-Reranker-4B`](https://huggingface.co/Qwen/Qwen3-Reranker-4B).
+
+Qwen 3 Reranker 4B in INT4. Pair with the Embedding kit for a fully local retrieval stack.
+
+| Property | Value |
+|---|---|
+| Base model | [Qwen/Qwen3-Reranker-4B](https://huggingface.co/Qwen/Qwen3-Reranker-4B) |
+| Quantization | INT4 weight-only |
+| Approx. on-disk size | ~2.7 GB |
+| Languages | English |
+
+## Load (vLLM)
+
+```bash
+vllm serve drawais/Qwen3-Reranker-4B-AWQ-INT4 \
+  --max-model-len 32768 \
+  --gpu-memory-utilization 0.94
+```
+
+```python
+from vllm import LLM, SamplingParams
+llm = LLM(model="drawais/Qwen3-Reranker-4B-AWQ-INT4", max_model_len=32768)
+print(llm.generate(["Hello!"], SamplingParams(max_tokens=128))[0].outputs[0].text)
+```
+
+## Footprint
+
+~2.7 GB on disk. Recommended VRAM: enough headroom for KV cache.
+
+## License
+
+This artifact follows the license of the base model: **apache-2.0**. See [`LICENSE`](LICENSE) for terms.
+
+## Acknowledgements
+
+Source model © its respective authors.

chat_template.jinja

@@ -0,0 +1,15 @@
+{%- set instruction = messages | selectattr("role", "eq", "system") | map(attribute="content") | first | default("Given a web search query, retrieve relevant passages that answer the query") -%}
+{%- set query_text = messages | selectattr("role", "eq", "query") | map(attribute="content") | first -%}
+{%- set document_text = messages | selectattr("role", "eq", "document") | map(attribute="content") | first -%}
+<|im_start|>system
+Judge whether the Document meets the requirements based on the Query and the Instruct provided. Note that the answer can only be "yes" or "no".<|im_end|>
+<|im_start|>user
+<Instruct>: {{ instruction }}
+<Query>: {{ query_text }}
+<Document>: {{ document_text }}<|im_end|>
+<|im_start|>assistant
+<think>
+
+</think>
+
+

config.json

@@ -0,0 +1,108 @@
+{
+  "architectures": [
+    "Qwen3ForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "bos_token_id": 151643,
+  "dtype": "bfloat16",
+  "eos_token_id": 151645,
+  "head_dim": 128,
+  "hidden_act": "silu",
+  "hidden_size": 2560,
+  "initializer_range": 0.02,
+  "intermediate_size": 9728,
+  "layer_types": [
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention"
+  ],
+  "max_position_embeddings": 40960,
+  "max_window_layers": 36,
+  "model_type": "qwen3",
+  "num_attention_heads": 32,
+  "num_hidden_layers": 36,
+  "num_key_value_heads": 8,
+  "pad_token_id": null,
+  "quantization_config": {
+    "config_groups": {
+      "group_0": {
+        "format": "pack-quantized",
+        "input_activations": null,
+        "output_activations": null,
+        "targets": [
+          "Linear"
+        ],
+        "weights": {
+          "actorder": null,
+          "block_structure": null,
+          "dynamic": false,
+          "group_size": 128,
+          "num_bits": 4,
+          "observer": "memoryless_minmax",
+          "observer_kwargs": {},
+          "scale_dtype": null,
+          "strategy": "group",
+          "symmetric": true,
+          "type": "int",
+          "zp_dtype": null
+        }
+      }
+    },
+    "format": "pack-quantized",
+    "global_compression_ratio": null,
+    "ignore": [
+      "lm_head"
+    ],
+    "kv_cache_scheme": null,
+    "quant_method": "compressed-tensors",
+    "quantization_status": "compressed",
+    "sparsity_config": {},
+    "transform_config": {},
+    "version": "0.15.1.a20260428"
+  },
+  "rms_norm_eps": 1e-06,
+  "rope_parameters": {
+    "rope_theta": 1000000,
+    "rope_type": "default"
+  },
+  "sliding_window": null,
+  "tie_word_embeddings": true,
+  "transformers_version": "5.8.0.dev0",
+  "use_cache": true,
+  "use_sliding_window": false,
+  "vocab_size": 151669
+}

generation_config.json

@@ -0,0 +1,13 @@
+{
+  "bos_token_id": 151643,
+  "do_sample": true,
+  "eos_token_id": [
+    151645,
+    151643
+  ],
+  "pad_token_id": 151643,
+  "temperature": 0.6,
+  "top_k": 20,
+  "top_p": 0.95,
+  "transformers_version": "5.8.0.dev0"
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e221de455aa1a946f7e0c7a457913973ddf3df921aa74750633d82fb74a255e0
+size 2650472528

recipe.yaml

@@ -0,0 +1,23 @@
+default_stage:
+  default_modifiers:
+    AWQModifier:
+      mappings:
+      - smooth_layer: re:.*input_layernorm$
+        balance_layers: ['re:.*q_proj$', 're:.*k_proj$', 're:.*v_proj$']
+        activation_hook_target: null
+      - smooth_layer: re:.*v_proj$
+        balance_layers: ['re:.*o_proj$']
+        activation_hook_target: null
+      - smooth_layer: re:.*post_attention_layernorm$
+        balance_layers: ['re:.*gate_proj$', 're:.*up_proj$']
+        activation_hook_target: null
+      - smooth_layer: re:.*up_proj$
+        balance_layers: ['re:.*down_proj$']
+        activation_hook_target: null
+      duo_scaling: true
+      n_grid: 20
+    QuantizationModifier:
+      targets: [Linear]
+      ignore: [lm_head, 're:.*embed.*', 're:.*router.*', 're:.*\.gate$']
+      scheme: W4A16
+      bypass_divisibility_checks: false

tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e946ac23b6a68f7a2abbe7b3c22190673c6d3d159b85305268db51b2729ac68a
+size 11422749

tokenizer_config.json

@@ -0,0 +1,15 @@
+{
+  "add_prefix_space": false,
+  "backend": "tokenizers",
+  "bos_token": null,
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "errors": "replace",
+  "is_local": false,
+  "local_files_only": false,
+  "model_max_length": 131072,
+  "pad_token": "<|endoftext|>",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}