File size: 13,763 Bytes
a92633c
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
9f3578f
a92633c
 
 
5c09627
a92633c
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
de1686b
 
a92633c
ade73f1
 
deada0d
 
47a64ba
 
 
 
deada0d
 
7a5f8e0
 
ade73f1
a92633c
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
25e781a
a92633c
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
---
quantized_by: jartine
model_creator: ibm-granite
pipeline_tag: text-generation
base_model: ibm-granite/granite-34b-code-base
inference: true
license: apache-2.0
datasets:
- bigcode/commitpackft
- TIGER-Lab/MathInstruct
- meta-math/MetaMathQA
- glaiveai/glaive-code-assistant-v3
- glaive-function-calling-v2
- bugdaryan/sql-create-context-instruction
- garage-bAInd/Open-Platypus
- nvidia/HelpSteer
metrics:
- code_eval
library_name: transformers
tags:
- code
- granite
model-index:
- name: granite-34b-code-instruct
  results:
  - task:
      type: text-generation
    dataset:
        type: bigcode/humanevalpack 
        name: HumanEvalSynthesis(Python)
    metrics:
    - name: pass@1
      type: pass@1
      value: 62.2
      veriefied: false
  - task:
      type: text-generation
    dataset:
        type: bigcode/humanevalpack  
        name: HumanEvalSynthesis(JavaScript)
    metrics:
    - name: pass@1
      type: pass@1
      value: 56.7
      veriefied: false
  - task:
      type: text-generation
    dataset:
        type: bigcode/humanevalpack  
        name: HumanEvalSynthesis(Java)
    metrics:
    - name: pass@1
      type: pass@1
      value: 62.8
      veriefied: false
  - task:
      type: text-generation
    dataset:
        type: bigcode/humanevalpack  
        name: HumanEvalSynthesis(Go)
    metrics:
    - name: pass@1
      type: pass@1
      value: 47.6
      veriefied: false
  - task:
      type: text-generation
    dataset:
        type: bigcode/humanevalpack  
        name: HumanEvalSynthesis(C++)
    metrics:
    - name: pass@1
      type: pass@1
      value: 57.9
      veriefied: false
  - task:
      type: text-generation
    dataset:
        type: bigcode/humanevalpack  
        name: HumanEvalSynthesis(Rust)
    metrics:
    - name: pass@1
      type: pass@1
      value: 41.5
      veriefied: false
  - task:
      type: text-generation
    dataset:
        type: bigcode/humanevalpack  
        name:  HumanEvalExplain(Python)
    metrics:
    - name: pass@1
      type: pass@1
      value: 53.0
      veriefied: false
  - task:
      type: text-generation
    dataset:
        type: bigcode/humanevalpack  
        name:  HumanEvalExplain(JavaScript)
    metrics:
    - name: pass@1
      type: pass@1
      value: 45.1
      veriefied: false
  - task:
      type: text-generation
    dataset:
        type: bigcode/humanevalpack  
        name:  HumanEvalExplain(Java)
    metrics:
    - name: pass@1
      type: pass@1
      value: 50.6
      veriefied: false
  - task:
      type: text-generation
    dataset:
        type: bigcode/humanevalpack  
        name:  HumanEvalExplain(Go)
    metrics:
    - name: pass@1
      type: pass@1
      value: 36.0
      veriefied: false
  - task:
      type: text-generation
    dataset:
        type: bigcode/humanevalpack  
        name:  HumanEvalExplain(C++)
    metrics:
    - name: pass@1
      type: pass@1
      value: 42.7
      veriefied: false
  - task:
      type: text-generation
    dataset:
        type: bigcode/humanevalpack  
        name:  HumanEvalExplain(Rust)
    metrics:
    - name: pass@1
      type: pass@1
      value: 23.8
      veriefied: false
  - task:
      type: text-generation
    dataset:
        type: bigcode/humanevalpack  
        name:  HumanEvalFix(Python)
    metrics:
    - name: pass@1
      type: pass@1
      value: 54.9
      veriefied: false
  - task:
      type: text-generation
    dataset:
        type: bigcode/humanevalpack  
        name:  HumanEvalFix(JavaScript)
    metrics:
    - name: pass@1
      type: pass@1
      value: 47.6
      veriefied: false
  - task:
      type: text-generation
    dataset:
        type: bigcode/humanevalpack  
        name:  HumanEvalFix(Java)
    metrics:
    - name: pass@1
      type: pass@1
      value: 55.5
      veriefied: false
  - task:
      type: text-generation
    dataset:
        type: bigcode/humanevalpack  
        name:  HumanEvalFix(Go)
    metrics:
    - name: pass@1
      type: pass@1
      value: 51.2
      veriefied: false
  - task:
      type: text-generation
    dataset:
        type: bigcode/humanevalpack  
        name:  HumanEvalFix(C++)
    metrics:
    - name: pass@1
      type: pass@1
      value: 47.0
      veriefied: false
  - task:
      type: text-generation
    dataset:
        type: bigcode/humanevalpack  
        name:  HumanEvalFix(Rust)
    metrics:
    - name: pass@1
      type: pass@1
      value: 45.1
      veriefied: false
---

# Granite 34B Code Instruct - llamafile

This repository contains executable weights (which we call
[llamafiles](https://github.com/Mozilla-Ocho/llamafile)) that run on
Linux, MacOS, Windows, FreeBSD, OpenBSD, and NetBSD for AMD64 and ARM64.

- Model creator: [IBM](https://hf.co/ibm-granite)
- Original model: [ibm-granite/granite-34b-code-instruct](https://huggingface.co/ibm-granite/granite-34b-code-instruct)
- Base model: [ibm-granite/granite-34b-code-base](https://huggingface.co/ibm-granite/granite-34b-code-base)

Granite 34B is a coding model released by IBM in April of 2024.

## Quickstart

Assuming your system has at least 64GB of RAM, you can try running the
following command which download, concatenate, and execute the model.

```
wget https://huggingface.co/jartine/granite-34b-code-instruct-llamafile/resolve/main/granite-34b-code-instruct.Q5_0.llamafile
chmod +x granite-34b-code-instruct.Q5_0.llamafile
./granite-34b-code-instruct.Q5_0.llamafile --help   # view manual
./granite-34b-code-instruct.Q5_0.llamafile          # launch web gui + oai api
./granite-34b-code-instruct.Q5_0.llamafile -p ...   # cli interface (scriptable)
```

Alternatively, you may download an official `llamafile` executable from
Mozilla Ocho on GitHub, in which case you can use the Granite llamafiles
as a simple weights data file.

```
llamafile -m granite-34b-code-instruct.Q5_0.llamafile ...
```

For further information, please see the [llamafile
README](https://github.com/mozilla-ocho/llamafile/).

Having **trouble?** See the ["Gotchas"
section](https://github.com/mozilla-ocho/llamafile/?tab=readme-ov-file#gotchas)
of the README.

## Prompting

The chat template is stored in the GGUF files. From the CLI interface,
Mistral style prompts seem to work with this model too:

```
[INST] {{prompt}} [/INST]
```

Command template:

```
./granite-34b-code-instruct.Q5_0.llamafile -p "[INST]{{prompt}}[/INST]"
```

The maximum context size of this model is 8192 tokens. These llamafiles
use a default context size of 512 tokens. Whenever you need the maximum
context size to be available with llamafile for any given model, you can
pass the `-c 0` flag. The default temperature for these llamafiles is 0.
It can be changed, e.g. `--temp 0.8`.

## Benchmarks

| hardware                                   | model\_filename                          | size       | test          | t/s             |
| :----------------------------------------- | :--------------------------------------- | ---------: | ------------: | --------------: |
| Apple M2 Ultra (60-core Metal GPU)         | granite-34b-code-instruct.Q5\_0          | 22.03 GiB  | pp512         | 159.02          |
| Apple M2 Ultra (60-core Metal GPU)         | granite-34b-code-instruct.Q5\_0          | 22.03 GiB  | tg16          | 15.39           |
| Apple M2 Ultra (60-core Metal GPU)         | granite-34b-code-instruct.Q8\_0          | 33.82 GiB  | pp512         | 186.14          |
| Apple M2 Ultra (60-core Metal GPU)         | granite-34b-code-instruct.Q8\_0          | 33.82 GiB  | tg16          | 14.13           |
| AMD Ryzen Threadripper PRO 7995WX (znver4) | granite-34b-code-instruct.Q5\_0          | 22.03 GiB  | pp512         | 95.08           |
| AMD Ryzen Threadripper PRO 7995WX (znver4) | granite-34b-code-instruct.Q5\_0          | 22.03 GiB  | tg16          | 7.78            |
| AMD Ryzen Threadripper PRO 7995WX (znver4) | granite-34b-code-instruct.Q8\_0          | 33.82 GiB  | pp512         | 94.34           |
| AMD Ryzen Threadripper PRO 7995WX (znver4) | granite-34b-code-instruct.Q8\_0          | 33.82 GiB  | tg16          | 5.61            |

## About Quantization

Our own evaluation of this model leads us to believe that it works best
with the `Q5_0` and `Q8_0` quants. We tried other quantization formats
such as `Q6_K` but it didn't seem to be a good of a fit for this model.

## About llamafile

llamafile is a new format introduced by Mozilla Ocho on Nov 20th 2023.
It uses Cosmopolitan Libc to turn LLM weights into runnable llama.cpp
binaries that run on the stock installs of six OSes for both ARM64 and
AMD64.

In addition to being executables, llamafiles are also zip archives. Each
llamafile contains a GGUF file, which you can extract using the `unzip`
command. If you want to change or add files to your llamafiles, then the
`zipalign` command (distributed on the llamafile github) should be used
instead of the traditional `zip` command.

---

![image/png](https://cdn-uploads.huggingface.co/production/uploads/62cd5057674cdb524450093d/1hzxoPwqkBJXshKVVe6_9.png)

# Granite-34B-Code-Instruct

## Model Summary
**Granite-34B-Code-Instruct** is a 34B parameter model fine tuned from *Granite-34B-Code-Base* on a combination of **permissively licensed** instruction data to enhance instruction following capabilities including logical reasoning and problem-solving skills.

- **Developers:** IBM Research
- **GitHub Repository:** [ibm-granite/granite-code-models](https://github.com/ibm-granite/granite-code-models)
- **Paper:** [Granite Code Models: A Family of Open Foundation Models for Code Intelligence](https://arxiv.org/abs/2405.04324)
- **Release Date**: May 6th, 2024
- **License:** [Apache 2.0](https://www.apache.org/licenses/LICENSE-2.0).

## Usage
### Intended use
The model is designed to respond to coding related instructions and can be used to build coding assistants.

<!-- TO DO: Check starcoder2 instruct code example that includes the template https://huggingface.co/bigcode/starcoder2-15b-instruct-v0.1 -->

### Generation
This is a simple example of how to use **Granite-34B-Code-Instruct** model.

```python
import torch
from transformers import AutoModelForCausalLM, AutoTokenizer
device = "cuda" # or "cpu"
model_path = "ibm-granite/granite-34b-code-instruct"
tokenizer = AutoTokenizer.from_pretrained(model_path)
# drop device_map if running on CPU
model = AutoModelForCausalLM.from_pretrained(model_path, device_map=device)
model.eval()
# change input text as desired
chat = [
    { "role": "user", "content": "Write a code to find the maximum value in a list of numbers." },
]
chat = tokenizer.apply_chat_template(chat, tokenize=False, add_generation_prompt=True)
# tokenize the text
input_tokens = tokenizer(chat, return_tensors="pt")
# transfer tokenized inputs to the device
for i in input_tokens:
    input_tokens[i] = input_tokens[i].to(device)
# generate output tokens
output = model.generate(**input_tokens, max_new_tokens=100)
# decode output tokens into text
output = tokenizer.batch_decode(output)
# loop over the batch to print, in this example the batch size is 1
for i in output:
    print(i)
```


<!-- TO DO: Check this part -->
## Training Data
Granite Code Instruct models are trained on the following types of data.
* Code Commits Datasets: we sourced code commits data from the [CommitPackFT](https://huggingface.co/datasets/bigcode/commitpackft) dataset, a filtered version of the full CommitPack dataset. From CommitPackFT dataset, we only consider data for 92 programming languages. Our inclusion criteria boils down to selecting programming languages common across CommitPackFT and the 116 languages that we considered to pretrain the code-base model (*Granite-34B-Code-Base*). 
* Math Datasets: We consider two high-quality math datasets, [MathInstruct](https://huggingface.co/datasets/TIGER-Lab/MathInstruct) and [MetaMathQA](https://huggingface.co/datasets/meta-math/MetaMathQA). Due to license issues, we filtered out GSM8K-RFT and Camel-Math from MathInstruct dataset. 
* Code Instruction Datasets: We use [Glaive-Code-Assistant-v3](https://huggingface.co/datasets/glaiveai/glaive-code-assistant-v3), [Glaive-Function-Calling-v2](https://huggingface.co/datasets/glaiveai/glaive-function-calling-v2), [NL2SQL11](https://huggingface.co/datasets/bugdaryan/sql-create-context-instruction) and a small collection of synthetic API calling datasets.
* Language Instruction Datasets: We include high-quality datasets such as [HelpSteer](https://huggingface.co/datasets/nvidia/HelpSteer) and an open license-filtered version of [Platypus](https://huggingface.co/datasets/garage-bAInd/Open-Platypus). We also include a collection of hardcoded prompts to ensure our model generates correct outputs given inquiries about its name or developers.

## Infrastructure
We train the Granite Code models using two of IBM's super computing clusters, namely Vela and Blue Vela, both outfitted with NVIDIA A100 and H100 GPUs respectively. These clusters provide a scalable and efficient infrastructure for training our models over thousands of GPUs.

## Ethical Considerations and Limitations
Granite code instruct models are primarily finetuned using instruction-response pairs across a specific set of programming languages. Thus, their performance may be limited with out-of-domain programming languages. In this situation, it is beneficial providing few-shot examples to steer the model's output. Moreover, developers should perform safety testing and target-specific tuning before deploying these models on critical applications. The model also inherits ethical considerations and limitations from its base model. For more information, please refer to *[Granite-34B-Code-Base](https://huggingface.co/ibm-granite/granite-34b-code-base)* model card.