Hugging Face's logo

strykes
/
SteraVibeThinker

Text Generation
Transformers
Safetensors
GGUF
code
agent
tool-use
qwen2
conversational
Model card Files
xet
 Community
strykes commited on
Commit
028191d
·
verified ·
1 Parent(s): df702ae

Add files using upload-large-folder tool

Browse files
Files changed (11) hide show
  1. .gitattributes +3 -34
  2. README.md +60 -0
  3. SteraVibeThinker-Q4_K_M.gguf +3 -0
  4. raw_weights/chat_template.jinja +54 -0
  5. raw_weights/config.json +69 -0
  6. raw_weights/generation_config.json +8 -0
  7. raw_weights/model.safetensors +3 -0
  8. raw_weights/tokenizer.json +3 -0
  9. raw_weights/tokenizer_config.json +16 -0
  10. raw_weights/training_args.bin +3 -0
  11. val_meta.jsonl +0 -0
.gitattributes CHANGED
@@ -1,35 +1,4 @@
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
 
1
+ *.gguf filter=lfs diff=lfs merge=lfs -text
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
2
  *.safetensors filter=lfs diff=lfs merge=lfs -text
3
+ raw_weights/training_args.bin filter=lfs diff=lfs merge=lfs -text
4
+ raw_weights/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 
 
 
 
 
 
 
 
README.md ADDED
@@ -0,0 +1,60 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: mit
3
+ base_model: WeiboAI/VibeThinker-3B
4
+ tags:
5
+ - code
6
+ - agent
7
+ - tool-use
8
+ - gguf
9
+ - qwen2
10
+ library_name: transformers
11
+ pipeline_tag: text-generation
12
+ ---
13
+
14
+ # SteraVibeThinker
15
+
16
+ A full fine-tune of [WeiboAI/VibeThinker-3B](https://huggingface.co/WeiboAI/VibeThinker-3B)
17
+ (a 3B reasoning model built on the Qwen2.5-3B / Qwen2.5-Coder-3B architecture) on
18
+ the ~30k-example **Tiny-Giant** agentic tool-use dataset.
19
+
20
+ The goal: keep VibeThinker's strong verifiable-reasoning core while teaching it the
21
+ deterministic, Hermes/ChatML-style `<tool_call>` agent format used by the
22
+ Tiny-Giant harness.
23
+
24
+ ## Files
25
+
26
+ | File | Description |
27
+ |---|---|
28
+ | `SteraVibeThinker-Q4_K_M.gguf` | Q4_K_M quantization (~1.8 GB) — for `llama.cpp` / Ollama / LM Studio |
29
+ | `raw_weights/` | Full bf16 safetensors HF checkpoint (re-quantize to any GGUF level from here) |
30
+ | `val_meta.jsonl` | Held-out validation set shipped with the model |
31
+
32
+ ## Training
33
+
34
+ - **Base:** `WeiboAI/VibeThinker-3B` (MIT, Qwen2.5-3B architecture, ChatML-native)
35
+ - **Method:** full fine-tune (not LoRA), bf16 + gradient checkpointing
36
+ - **Data:** ~30k Tiny-Giant agentic tool-use conversations
37
+ - **Epochs:** 2 · **LR:** 7e-6 (cosine, 3% warmup) · **Seq len:** 4096
38
+ - **Loss:** full-sequence (tool results modeled as in-distribution context)
39
+
40
+ ## Prompt format
41
+
42
+ This model was trained with an **explicit ChatML / Hermes renderer**, not
43
+ `tokenizer.apply_chat_template`. Pin the ChatML template explicitly when serving —
44
+ do not rely on auto-detection. Tool calls use:
45
+
46
+ ```
47
+ <tool_call>
48
+ {"name": "<function-name>", "arguments": {...}}
49
+ </tool_call>
50
+ ```
51
+
52
+ ## Inference (llama.cpp)
53
+
54
+ ```bash
55
+ llama-cli -m SteraVibeThinker-Q4_K_M.gguf --chat-template chatml
56
+ ```
57
+
58
+ ## License
59
+
60
+ MIT, inherited from the VibeThinker-3B base model.
SteraVibeThinker-Q4_K_M.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0b26426e293bdc609d53d120b1b36ebf44183844d471ce1d716486dc09a48088
3
+ size 1929902176
raw_weights/chat_template.jinja ADDED
@@ -0,0 +1,54 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {%- if tools %}
2
+ {{- '<|im_start|>system\n' }}
3
+ {%- if messages[0]['role'] == 'system' %}
4
+ {{- messages[0]['content'] }}
5
+ {%- else %}
6
+ {{- 'You are a helpful assistant.' }}
7
+ {%- endif %}
8
+ {{- "\n\n# 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>" }}
9
+ {%- for tool in tools %}
10
+ {{- "\n" }}
11
+ {{- tool | tojson }}
12
+ {%- endfor %}
13
+ {{- "\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" }}
14
+ {%- else %}
15
+ {%- if messages[0]['role'] == 'system' %}
16
+ {{- '<|im_start|>system\n' + messages[0]['content'] + '<|im_end|>\n' }}
17
+ {%- else %}
18
+ {{- '<|im_start|>system\nYou are a helpful assistant.<|im_end|>\n' }}
19
+ {%- endif %}
20
+ {%- endif %}
21
+ {%- for message in messages %}
22
+ {%- if (message.role == "user") or (message.role == "system" and not loop.first) or (message.role == "assistant" and not message.tool_calls) %}
23
+ {{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>' + '\n' }}
24
+ {%- elif message.role == "assistant" %}
25
+ {{- '<|im_start|>' + message.role }}
26
+ {%- if message.content %}
27
+ {{- '\n' + message.content }}
28
+ {%- endif %}
29
+ {%- for tool_call in message.tool_calls %}
30
+ {%- if tool_call.function is defined %}
31
+ {%- set tool_call = tool_call.function %}
32
+ {%- endif %}
33
+ {{- '\n<tool_call>\n{"name": "' }}
34
+ {{- tool_call.name }}
35
+ {{- '", "arguments": ' }}
36
+ {{- tool_call.arguments | tojson }}
37
+ {{- '}\n</tool_call>' }}
38
+ {%- endfor %}
39
+ {{- '<|im_end|>\n' }}
40
+ {%- elif message.role == "tool" %}
41
+ {%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != "tool") %}
42
+ {{- '<|im_start|>user' }}
43
+ {%- endif %}
44
+ {{- '\n<tool_response>\n' }}
45
+ {{- message.content }}
46
+ {{- '\n</tool_response>' }}
47
+ {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
48
+ {{- '<|im_end|>\n' }}
49
+ {%- endif %}
50
+ {%- endif %}
51
+ {%- endfor %}
52
+ {%- if add_generation_prompt %}
53
+ {{- '<|im_start|>assistant\n' }}
54
+ {%- endif %}
raw_weights/config.json ADDED
@@ -0,0 +1,69 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "Qwen2ForCausalLM"
4
+ ],
5
+ "attention_dropout": 0.0,
6
+ "bos_token_id": null,
7
+ "dtype": "bfloat16",
8
+ "eos_token_id": 151643,
9
+ "hidden_act": "silu",
10
+ "hidden_size": 2048,
11
+ "initializer_range": 0.02,
12
+ "intermediate_size": 11008,
13
+ "layer_types": [
14
+ "full_attention",
15
+ "full_attention",
16
+ "full_attention",
17
+ "full_attention",
18
+ "full_attention",
19
+ "full_attention",
20
+ "full_attention",
21
+ "full_attention",
22
+ "full_attention",
23
+ "full_attention",
24
+ "full_attention",
25
+ "full_attention",
26
+ "full_attention",
27
+ "full_attention",
28
+ "full_attention",
29
+ "full_attention",
30
+ "full_attention",
31
+ "full_attention",
32
+ "full_attention",
33
+ "full_attention",
34
+ "full_attention",
35
+ "full_attention",
36
+ "full_attention",
37
+ "full_attention",
38
+ "full_attention",
39
+ "full_attention",
40
+ "full_attention",
41
+ "full_attention",
42
+ "full_attention",
43
+ "full_attention",
44
+ "full_attention",
45
+ "full_attention",
46
+ "full_attention",
47
+ "full_attention",
48
+ "full_attention",
49
+ "full_attention"
50
+ ],
51
+ "max_position_embeddings": 131072,
52
+ "max_window_layers": 36,
53
+ "model_type": "qwen2",
54
+ "num_attention_heads": 16,
55
+ "num_hidden_layers": 36,
56
+ "num_key_value_heads": 2,
57
+ "pad_token_id": 151643,
58
+ "rms_norm_eps": 1e-06,
59
+ "rope_parameters": {
60
+ "rope_theta": 1000000.0,
61
+ "rope_type": "default"
62
+ },
63
+ "sliding_window": null,
64
+ "tie_word_embeddings": true,
65
+ "transformers_version": "5.12.1",
66
+ "use_cache": false,
67
+ "use_sliding_window": false,
68
+ "vocab_size": 151936
69
+ }
raw_weights/generation_config.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "eos_token_id": [
3
+ 151643
4
+ ],
5
+ "max_new_tokens": 2048,
6
+ "pad_token_id": 151643,
7
+ "transformers_version": "5.12.1"
8
+ }
raw_weights/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5e20a821f0a4b43c7e0797b34dd6956a5e7dbc22c8a118f74e0aeabeeb1fc8c5
3
+ size 6171927112
raw_weights/tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b17e16899b7fab7e695509f84bac5f10ed12804f0a590e935941e5af7f092f7f
3
+ size 11422263
raw_weights/tokenizer_config.json ADDED
@@ -0,0 +1,16 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": false,
3
+ "backend": "tokenizers",
4
+ "bos_token": null,
5
+ "clean_up_tokenization_spaces": false,
6
+ "eos_token": "<|endoftext|>",
7
+ "errors": "replace",
8
+ "is_local": false,
9
+ "local_files_only": false,
10
+ "model_max_length": 131072,
11
+ "pad_token": "<|endoftext|>",
12
+ "padding_side": "right",
13
+ "split_special_tokens": false,
14
+ "tokenizer_class": "Qwen2Tokenizer",
15
+ "unk_token": null
16
+ }
raw_weights/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2c45d3b519956d21d4fc74be024458e2a5cd67ddbe7df49fa1e5c0012157dfe1
3
+ size 5176
val_meta.jsonl ADDED
The diff for this file is too large to render. See raw diff