lballore commited on
Commit
bae95e0
·
0 Parent(s):

First release of llimba-demo

Browse files
Files changed (5) hide show
  1. .gitattributes +35 -0
  2. LICENSE +201 -0
  3. README.md +84 -0
  4. app.py +258 -0
  5. requirements.txt +5 -0
.gitattributes ADDED
@@ -0,0 +1,35 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
LICENSE ADDED
@@ -0,0 +1,201 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ Apache License
2
+ Version 2.0, January 2004
3
+ http://www.apache.org/licenses/
4
+
5
+ TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
6
+
7
+ 1. Definitions.
8
+
9
+ "License" shall mean the terms and conditions for use, reproduction,
10
+ and distribution as defined by Sections 1 through 9 of this document.
11
+
12
+ "Licensor" shall mean the copyright owner or entity authorized by
13
+ the copyright owner that is granting the License.
14
+
15
+ "Legal Entity" shall mean the union of the acting entity and all
16
+ other entities that control, are controlled by, or are under common
17
+ control with that entity. For the purposes of this definition,
18
+ "control" means (i) the power, direct or indirect, to cause the
19
+ direction or management of such entity, whether by contract or
20
+ otherwise, or (ii) ownership of fifty percent (50%) or more of the
21
+ outstanding shares, or (iii) beneficial ownership of such entity.
22
+
23
+ "You" (or "Your") shall mean an individual or Legal Entity
24
+ exercising permissions granted by this License.
25
+
26
+ "Source" form shall mean the preferred form for making modifications,
27
+ including but not limited to software source code, documentation
28
+ source, and configuration files.
29
+
30
+ "Object" form shall mean any form resulting from mechanical
31
+ transformation or translation of a Source form, including but
32
+ not limited to compiled object code, generated documentation,
33
+ and conversions to other media types.
34
+
35
+ "Work" shall mean the work of authorship, whether in Source or
36
+ Object form, made available under the License, as indicated by a
37
+ copyright notice that is included in or attached to the work
38
+ (an example is provided in the Appendix below).
39
+
40
+ "Derivative Works" shall mean any work, whether in Source or Object
41
+ form, that is based on (or derived from) the Work and for which the
42
+ editorial revisions, annotations, elaborations, or other modifications
43
+ represent, as a whole, an original work of authorship. For the purposes
44
+ of this License, Derivative Works shall not include works that remain
45
+ separable from, or merely link (or bind by name) to the interfaces of,
46
+ the Work and Derivative Works thereof.
47
+
48
+ "Contribution" shall mean any work of authorship, including
49
+ the original version of the Work and any modifications or additions
50
+ to that Work or Derivative Works thereof, that is intentionally
51
+ submitted to Licensor for inclusion in the Work by the copyright owner
52
+ or by an individual or Legal Entity authorized to submit on behalf of
53
+ the copyright owner. For the purposes of this definition, "submitted"
54
+ means any form of electronic, verbal, or written communication sent
55
+ to the Licensor or its representatives, including but not limited to
56
+ communication on electronic mailing lists, source code control systems,
57
+ and issue tracking systems that are managed by, or on behalf of, the
58
+ Licensor for the purpose of discussing and improving the Work, but
59
+ excluding communication that is conspicuously marked or otherwise
60
+ designated in writing by the copyright owner as "Not a Contribution."
61
+
62
+ "Contributor" shall mean Licensor and any individual or Legal Entity
63
+ on behalf of whom a Contribution has been received by Licensor and
64
+ subsequently incorporated within the Work.
65
+
66
+ 2. Grant of Copyright License. Subject to the terms and conditions of
67
+ this License, each Contributor hereby grants to You a perpetual,
68
+ worldwide, non-exclusive, no-charge, royalty-free, irrevocable
69
+ copyright license to reproduce, prepare Derivative Works of,
70
+ publicly display, publicly perform, sublicense, and distribute the
71
+ Work and such Derivative Works in Source or Object form.
72
+
73
+ 3. Grant of Patent License. Subject to the terms and conditions of
74
+ this License, each Contributor hereby grants to You a perpetual,
75
+ worldwide, non-exclusive, no-charge, royalty-free, irrevocable
76
+ (except as stated in this section) patent license to make, have made,
77
+ use, offer to sell, sell, import, and otherwise transfer the Work,
78
+ where such license applies only to those patent claims licensable
79
+ by such Contributor that are necessarily infringed by their
80
+ Contribution(s) alone or by combination of their Contribution(s)
81
+ with the Work to which such Contribution(s) was submitted. If You
82
+ institute patent litigation against any entity (including a
83
+ cross-claim or counterclaim in a lawsuit) alleging that the Work
84
+ or a Contribution incorporated within the Work constitutes direct
85
+ or contributory patent infringement, then any patent licenses
86
+ granted to You under this License for that Work shall terminate
87
+ as of the date such litigation is filed.
88
+
89
+ 4. Redistribution. You may reproduce and distribute copies of the
90
+ Work or Derivative Works thereof in any medium, with or without
91
+ modifications, and in Source or Object form, provided that You
92
+ meet the following conditions:
93
+
94
+ (a) You must give any other recipients of the Work or
95
+ Derivative Works a copy of this License; and
96
+
97
+ (b) You must cause any modified files to carry prominent notices
98
+ stating that You changed the files; and
99
+
100
+ (c) You must retain, in the Source form of any Derivative Works
101
+ that You distribute, all copyright, patent, trademark, and
102
+ attribution notices from the Source form of the Work,
103
+ excluding those notices that do not pertain to any part of
104
+ the Derivative Works; and
105
+
106
+ (d) If the Work includes a "NOTICE" text file as part of its
107
+ distribution, then any Derivative Works that You distribute must
108
+ include a readable copy of the attribution notices contained
109
+ within such NOTICE file, excluding those notices that do not
110
+ pertain to any part of the Derivative Works, in at least one
111
+ of the following places: within a NOTICE text file distributed
112
+ as part of the Derivative Works; within the Source form or
113
+ documentation, if provided along with the Derivative Works; or,
114
+ within a display generated by the Derivative Works, if and
115
+ wherever such third-party notices normally appear. The contents
116
+ of the NOTICE file are for informational purposes only and
117
+ do not modify the License. You may add Your own attribution
118
+ notices within Derivative Works that You distribute, alongside
119
+ or as an addendum to the NOTICE text from the Work, provided
120
+ that such additional attribution notices cannot be construed
121
+ as modifying the License.
122
+
123
+ You may add Your own copyright statement to Your modifications and
124
+ may provide additional or different license terms and conditions
125
+ for use, reproduction, or distribution of Your modifications, or
126
+ for any such Derivative Works as a whole, provided Your use,
127
+ reproduction, and distribution of the Work otherwise complies with
128
+ the conditions stated in this License.
129
+
130
+ 5. Submission of Contributions. Unless You explicitly state otherwise,
131
+ any Contribution intentionally submitted for inclusion in the Work
132
+ by You to the Licensor shall be under the terms and conditions of
133
+ this License, without any additional terms or conditions.
134
+ Notwithstanding the above, nothing herein shall supersede or modify
135
+ the terms of any separate license agreement you may have executed
136
+ with Licensor regarding such Contributions.
137
+
138
+ 6. Trademarks. This License does not grant permission to use the trade
139
+ names, trademarks, service marks, or product names of the Licensor,
140
+ except as required for reasonable and customary use in describing the
141
+ origin of the Work and reproducing the content of the NOTICE file.
142
+
143
+ 7. Disclaimer of Warranty. Unless required by applicable law or
144
+ agreed to in writing, Licensor provides the Work (and each
145
+ Contributor provides its Contributions) on an "AS IS" BASIS,
146
+ WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
147
+ implied, including, without limitation, any warranties or conditions
148
+ of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
149
+ PARTICULAR PURPOSE. You are solely responsible for determining the
150
+ appropriateness of using or redistributing the Work and assume any
151
+ risks associated with Your exercise of permissions under this License.
152
+
153
+ 8. Limitation of Liability. In no event and under no legal theory,
154
+ whether in tort (including negligence), contract, or otherwise,
155
+ unless required by applicable law (such as deliberate and grossly
156
+ negligent acts) or agreed to in writing, shall any Contributor be
157
+ liable to You for damages, including any direct, indirect, special,
158
+ incidental, or consequential damages of any character arising as a
159
+ result of this License or out of the use or inability to use the
160
+ Work (including but not limited to damages for loss of goodwill,
161
+ work stoppage, computer failure or malfunction, or any and all
162
+ other commercial damages or losses), even if such Contributor
163
+ has been advised of the possibility of such damages.
164
+
165
+ 9. Accepting Warranty or Additional Liability. While redistributing
166
+ the Work or Derivative Works thereof, You may choose to offer,
167
+ and charge a fee for, acceptance of support, warranty, indemnity,
168
+ or other liability obligations and/or rights consistent with this
169
+ License. However, in accepting such obligations, You may act only
170
+ on Your own behalf and on Your sole responsibility, not on behalf
171
+ of any other Contributor, and only if You agree to indemnify,
172
+ defend, and hold each Contributor harmless for any liability
173
+ incurred by, or claims asserted against, such Contributor by reason
174
+ of your accepting any such warranty or additional liability.
175
+
176
+ END OF TERMS AND CONDITIONS
177
+
178
+ APPENDIX: How to apply the Apache License to your work.
179
+
180
+ To apply the Apache License to your work, attach the following
181
+ boilerplate notice, with the fields enclosed by brackets "[]"
182
+ replaced with your own identifying information. (Don't include
183
+ the brackets!) The text should be enclosed in the appropriate
184
+ comment syntax for the file format. We also recommend that a
185
+ file or class name and description of purpose be included on the
186
+ same "printed page" as the copyright notice for easier
187
+ identification within third-party archives.
188
+
189
+ Copyright [yyyy] [name of copyright owner]
190
+
191
+ Licensed under the Apache License, Version 2.0 (the "License");
192
+ you may not use this file except in compliance with the License.
193
+ You may obtain a copy of the License at
194
+
195
+ http://www.apache.org/licenses/LICENSE-2.0
196
+
197
+ Unless required by applicable law or agreed to in writing, software
198
+ distributed under the License is distributed on an "AS IS" BASIS,
199
+ WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
200
+ See the License for the specific language governing permissions and
201
+ limitations under the License.
README.md ADDED
@@ -0,0 +1,84 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ title: LLiMba 3B Demo
3
+ emoji: 💬
4
+ colorFrom: yellow
5
+ colorTo: purple
6
+ sdk: gradio
7
+ sdk_version: 6.5.1
8
+ app_file: app.py
9
+ pinned: false
10
+ license: apache-2.0
11
+ short_description: Chat with LLiMba, an open 3B LLM that speaks Sardinian
12
+ hardware: zero-a10g
13
+ models:
14
+ - lballore/llimba-3b-instruct
15
+ tags:
16
+ - sardinian
17
+ - limba-sarda-comuna
18
+ - lsc
19
+ - low-resource
20
+ - endangered-language
21
+ - chat
22
+ - demo
23
+ ---
24
+
25
+ # 💬 LLiMba 3B Demo
26
+
27
+ A live demo of [**LLiMba-3B-Instruct**](https://huggingface.co/lballore/llimba-3b-instruct), an open 3B-parameter Sardinian-speaking language model adapted from Qwen2.5-3B-Instruct on a single consumer GPU.
28
+
29
+ LLiMba speaks fluent **Sardinian** (LSC, the standardized written form), accepts **Logudorese** and **Campidanese** input, and retains the multilingual capabilities of its Qwen2.5 base, so it also handles English, Italian, Spanish, and other Romance languages.
30
+
31
+ ## Try these prompts
32
+
33
+ If you don't speak Sardinian, here are starters that work well:
34
+
35
+ **Conversation:**
36
+ - *Salude! Comente ìstas?* (Hi! How are you?)
37
+
38
+ **Translation:**
39
+ - *Translate to Sardinian: "The Mediterranean is rough today."*
40
+ - *Traduzi in italianu: «Sa Sardigna est una ìsula bella meda in su Mediterràneu.»*
41
+
42
+ **Sardinian culture and history:**
43
+ - *Chie fiat Gigi Riva?* (Who was Gigi Riva?)
44
+ - *Ite est su «cantu a tenore» sardu?* (What is Sardinian *cantu a tenore*?)
45
+
46
+ **Open-ended:**
47
+ - *Iscrie unu paragrafu in sardu subra de sa Sardigna.* (Write a paragraph in Sardinian about Sardinia.)
48
+
49
+ ## Inference settings
50
+
51
+ The sidebar exposes the standard generation parameters. Defaults are tuned for natural Sardinian conversation; the model card recommends three preset profiles depending on use case:
52
+
53
+ | Use case | Temperature | Top-p | Top-k | Repetition penalty |
54
+ |---|---:|---:|---:|---:|
55
+ | Translation, factual Q&A | 0.01 (greedy) | 1.0 | 1 | 1.0 |
56
+ | Conversational chat (default) | 0.3 | 0.9 | 40 | 1.05 |
57
+ | Creative or long-form | 0.7 | 0.9 | 40 | 1.1 |
58
+
59
+ Temperatures above 0.7 may cause language-boundary drift (Sardinian to Italian) and amplify morphological hallucination on long open-ended prompts. The model was trained with Romance replay data to mitigate this, but the safe upper bound for production use is around 0.7.
60
+
61
+ ## About the project
62
+
63
+ LLiMba is an open project to bring Sardinian, an endangered Romance language with roughly one million speakers, into modern NLP. The full release includes:
64
+
65
+ - 🤖 **Model:** [lballore/llimba-3b-instruct](https://huggingface.co/lballore/llimba-3b-instruct)
66
+ - 🤖 **Intermediate checkpoint:** [lballore/llimba-3b-instruct-cpt](https://huggingface.co/lballore/llimba-3b-instruct-cpt) (post-CPT, pre-SFT, for researchers)
67
+ - 📚 **Pretraining corpus:** [lballore/llimba-corpus](https://huggingface.co/datasets/lballore/llimba-corpus) (~13.9M tokens)
68
+ - 📚 **SFT data:** [lballore/llimba-sft](https://huggingface.co/datasets/lballore/llimba-sft) (~14K instruction pairs)
69
+ - 📚 **Eval set:** [lballore/llimba-flores-srd-eval](https://huggingface.co/datasets/lballore/llimba-flores-srd-eval) (FLORES-200 subset)
70
+ - 💻 **Code:** [github.com/lballore/LLiMba](https://github.com/lballore/LLiMba)
71
+
72
+ The model was adapted via continued pretraining on Sardinian text (with Romance replay data to prevent language drift) followed by supervised fine-tuning with rsLoRA on instruction pairs. Full methodology and benchmarks in the [model card](https://huggingface.co/lballore/llimba-3b-instruct).
73
+
74
+ ## Limitations
75
+
76
+ This is a 3B-parameter model and like all small LLMs it can produce confident wrong answers, especially on biographical or historical specifics. Treat factual outputs with skepticism.
77
+
78
+ On long open-ended generation, the model may occasionally produce phonotactically valid but non-attested Sardinian words. Structured prompts ("List the three main causes") work more reliably than open-ended ones ("Tell me about X").
79
+
80
+ The model targets LSC and was reviewed by a single native speaker of the Nuorese variant. Logudorese and Campidanese input is handled, but speakers of those variants may find outputs skew toward the standardized form.
81
+
82
+ ## License
83
+
84
+ The demo code is released under [Apache 2.0](LICENSE). The model weights are released under [Apache 2.0](LICENSE).
app.py ADDED
@@ -0,0 +1,258 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ import os
2
+
3
+ import gradio as gr
4
+ import spaces
5
+ import torch
6
+ from threading import Thread
7
+ from transformers import AutoModelForCausalLM, AutoTokenizer, TextIteratorStreamer
8
+
9
+ MODEL_ID = "lballore/llimba-3b-instruct"
10
+ HF_TOKEN = os.environ.get("HF_TOKEN")
11
+
12
+ # ---------------------------------------------------------------------------
13
+ # System prompts
14
+ # ---------------------------------------------------------------------------
15
+
16
+ CHAT_SYSTEM_PROMPT = (
17
+ "Ses unu assistente chi chistionat in sardu (LSC). "
18
+ "Risponde in manera curtza, clara e pretzisa, chene repetire sa dimanda. "
19
+ "Si non connosches una risposta o non ses seguru, naralu in manera onesta "
20
+ "imbetzes de imbentare."
21
+ )
22
+
23
+ TRANSLATE_SYSTEM_TEMPLATE = (
24
+ "Ses unu tradutore espertu. Traduzi in {tgt} su testu chi sighit. "
25
+ "Su testu est in {src}. Risponde solu cun sa tradutzione, "
26
+ "sentza cummentos o ispiegatziones."
27
+ )
28
+
29
+ # Language names in Sardinian, used to fill the translation system prompt
30
+ LANGUAGES = {
31
+ "Sardinian (LSC)": "sardu",
32
+ "Italian": "italianu",
33
+ "English": "inglesu",
34
+ "Spanish": "ispagnolu",
35
+ }
36
+
37
+ # ---------------------------------------------------------------------------
38
+ # Examples
39
+ # ---------------------------------------------------------------------------
40
+
41
+ CHAT_EXAMPLES = [
42
+ "Salude! Comente ìstas?",
43
+ "Cale est sa capitale de sa Sardigna?",
44
+ "Chie fiat Gigi Riva?",
45
+ "Ite est su «cantu a tenore» sardu?",
46
+ "Iscrie unu paragrafu in sardu subra de sa Sardigna.",
47
+ ]
48
+
49
+ TRANSLATE_EXAMPLES = [
50
+ "The weather is rough today.",
51
+ "Buongiorno, come stai? È una bellissima giornata oggi.",
52
+ "La cultura sarda è ricca di tradizioni antiche.",
53
+ ]
54
+
55
+ # ---------------------------------------------------------------------------
56
+ # Model loading (once at startup; ZeroGPU keeps weights on CPU until generate)
57
+ # ---------------------------------------------------------------------------
58
+
59
+ tokenizer = AutoTokenizer.from_pretrained(MODEL_ID, token=HF_TOKEN)
60
+ model = AutoModelForCausalLM.from_pretrained(
61
+ MODEL_ID,
62
+ dtype=torch.bfloat16,
63
+ device_map="auto",
64
+ token=HF_TOKEN,
65
+ )
66
+
67
+ # ---------------------------------------------------------------------------
68
+ # Helpers
69
+ # ---------------------------------------------------------------------------
70
+
71
+ def _to_text(content):
72
+ """Normalize Gradio 6.x message content (may be list of dicts) to string."""
73
+ if isinstance(content, str):
74
+ return content
75
+ if isinstance(content, list):
76
+ return "".join(
77
+ part.get("text", "") if isinstance(part, dict) else str(part)
78
+ for part in content
79
+ )
80
+ return str(content)
81
+
82
+
83
+ def _normalize_history(history):
84
+ return [
85
+ {"role": m["role"], "content": _to_text(m["content"])}
86
+ for m in history
87
+ ]
88
+
89
+
90
+ def _stream_response(messages, max_tokens, temperature, top_p, top_k, rep_penalty):
91
+ """Run model.generate in a background thread and yield tokens as they arrive."""
92
+ inputs = tokenizer.apply_chat_template(
93
+ messages,
94
+ add_generation_prompt=True,
95
+ return_tensors="pt",
96
+ return_dict=True,
97
+ ).to(model.device)
98
+
99
+ streamer = TextIteratorStreamer(
100
+ tokenizer,
101
+ skip_prompt=True,
102
+ skip_special_tokens=True,
103
+ )
104
+
105
+ do_sample = temperature > 0.0
106
+ generation_kwargs = dict(
107
+ **inputs,
108
+ streamer=streamer,
109
+ max_new_tokens=max_tokens,
110
+ do_sample=do_sample,
111
+ temperature=temperature if do_sample else 1.0,
112
+ top_p=top_p if do_sample else 1.0,
113
+ top_k=top_k if do_sample else 50,
114
+ repetition_penalty=rep_penalty,
115
+ pad_token_id=tokenizer.eos_token_id,
116
+ )
117
+
118
+ thread = Thread(target=model.generate, kwargs=generation_kwargs)
119
+ thread.start()
120
+
121
+ response = ""
122
+ for new_text in streamer:
123
+ response += new_text
124
+ yield response
125
+
126
+ # ---------------------------------------------------------------------------
127
+ # Per-tab respond functions
128
+ # ---------------------------------------------------------------------------
129
+
130
+ @spaces.GPU(duration=60)
131
+ def respond_chat(
132
+ message, history,
133
+ system_message, max_tokens, temperature, top_p, top_k, rep_penalty,
134
+ ):
135
+ messages = [{"role": "system", "content": system_message}]
136
+ messages.extend(_normalize_history(history))
137
+ messages.append({"role": "user", "content": _to_text(message)})
138
+
139
+ yield from _stream_response(
140
+ messages, max_tokens, temperature, top_p, top_k, rep_penalty,
141
+ )
142
+
143
+
144
+ @spaces.GPU(duration=60)
145
+ def respond_translate(
146
+ message, history,
147
+ source_lang, target_lang,
148
+ max_tokens, temperature, top_p, top_k, rep_penalty,
149
+ ):
150
+ src = LANGUAGES[source_lang]
151
+ tgt = LANGUAGES[target_lang]
152
+ system_message = TRANSLATE_SYSTEM_TEMPLATE.format(src=src, tgt=tgt)
153
+
154
+ messages = [{"role": "system", "content": system_message}]
155
+ messages.extend(_normalize_history(history))
156
+ messages.append({"role": "user", "content": _to_text(message)})
157
+
158
+ yield from _stream_response(
159
+ messages, max_tokens, temperature, top_p, top_k, rep_penalty,
160
+ )
161
+
162
+ # ---------------------------------------------------------------------------
163
+ # UI
164
+ # ---------------------------------------------------------------------------
165
+
166
+ DESCRIPTION = (
167
+ "Chat with [LLiMba-3B-Instruct](https://huggingface.co/lballore/llimba-3b-instruct), "
168
+ "an open 3B LLM that speaks **Sardinian** (LSC, with Logudorese and Campidanese "
169
+ "accepted as input). The model retains the multilingual capabilities of its Qwen2.5 base."
170
+ )
171
+
172
+
173
+ with gr.Blocks(title="LLiMba 3B Demo") as demo:
174
+ gr.Markdown("# 💬 LLiMba 3B Demo")
175
+ gr.Markdown(DESCRIPTION)
176
+
177
+ with gr.Tabs():
178
+ # ----- Chat tab -----
179
+ with gr.Tab("💬 Chat"):
180
+ gr.ChatInterface(
181
+ respond_chat,
182
+ additional_inputs=[
183
+ gr.Textbox(
184
+ value=CHAT_SYSTEM_PROMPT,
185
+ label="System message",
186
+ info="Default tells the model to be concise and admit uncertainty.",
187
+ lines=4,
188
+ ),
189
+ gr.Slider(
190
+ minimum=1, maximum=2048, value=512, step=1,
191
+ label="Max new tokens",
192
+ ),
193
+ gr.Slider(
194
+ minimum=0.0, maximum=1.0, value=0.3, step=0.05,
195
+ label="Temperature",
196
+ info="0 = greedy. ≤0.5 recommended to limit hallucination and language drift.",
197
+ ),
198
+ gr.Slider(
199
+ minimum=0.05, maximum=1.0, value=0.9, step=0.05,
200
+ label="Top-p (nucleus sampling)",
201
+ ),
202
+ gr.Slider(
203
+ minimum=1, maximum=100, value=40, step=1,
204
+ label="Top-k",
205
+ ),
206
+ gr.Slider(
207
+ minimum=1.0, maximum=2.0, value=1.05, step=0.05,
208
+ label="Repetition penalty",
209
+ ),
210
+ ],
211
+ examples=[[p] for p in CHAT_EXAMPLES],
212
+ cache_examples=False,
213
+ )
214
+
215
+ # ----- Translate tab -----
216
+ with gr.Tab("🌐 Translate"):
217
+ gr.ChatInterface(
218
+ respond_translate,
219
+ additional_inputs=[
220
+ gr.Dropdown(
221
+ choices=list(LANGUAGES.keys()),
222
+ value="English",
223
+ label="Source language",
224
+ ),
225
+ gr.Dropdown(
226
+ choices=list(LANGUAGES.keys()),
227
+ value="Sardinian (LSC)",
228
+ label="Target language",
229
+ ),
230
+ gr.Slider(
231
+ minimum=1, maximum=2048, value=512, step=1,
232
+ label="Max new tokens",
233
+ ),
234
+ gr.Slider(
235
+ minimum=0.0, maximum=1.0, value=0.0, step=0.05,
236
+ label="Temperature",
237
+ info="0 = greedy. Recommended for translation.",
238
+ ),
239
+ gr.Slider(
240
+ minimum=0.05, maximum=1.0, value=1.0, step=0.05,
241
+ label="Top-p",
242
+ ),
243
+ gr.Slider(
244
+ minimum=1, maximum=100, value=1, step=1,
245
+ label="Top-k",
246
+ ),
247
+ gr.Slider(
248
+ minimum=1.0, maximum=2.0, value=1.0, step=0.05,
249
+ label="Repetition penalty",
250
+ ),
251
+ ],
252
+ examples=[[p] for p in TRANSLATE_EXAMPLES],
253
+ cache_examples=False,
254
+ )
255
+
256
+
257
+ if __name__ == "__main__":
258
+ demo.launch(theme=gr.themes.Soft())
requirements.txt ADDED
@@ -0,0 +1,5 @@
 
 
 
 
 
 
1
+ gradio>=6.5.1
2
+ spaces
3
+ transformers>=4.46.0
4
+ torch
5
+ accelerate