beomi / Llama-3-KoEn-8B-xtuner-llava-preview

huggingface.co
Total runs: 86
24-hour runs: 4
7-day runs: 12
30-day runs: 67
Model's Last Updated: May 08 2024
image-text-to-text

Introduction of Llama-3-KoEn-8B-xtuner-llava-preview

Model Details of Llama-3-KoEn-8B-xtuner-llava-preview

Llama-3-KoEn-8B-xtuner-llava-preview ๐ŸŒ‹

Llama-3-KoEn-8B-xtuner-llava-preview ๐ŸŒ‹ is Korean based MutliModal based on Llava architecture, merged with ChatVector methods leveraging 2 models:

  1. beomi/Llama-3-KoEn-8B-preview
  2. xtuner/llava-llama-3-8b-transformers
Model Details
Model Description
Direct Use

Cat walking on frozen Han-River, Seoul

Two version recommended

v1. revision='a38aac3' : Basic ChatVector, with 25B+ trained KoEn ckpt(rev. d4d25a2) .

v1-1. revision='0224971' : Basic ChatVector, with 40B+ trained KoEn ckpt(rev. ad39b32) .

v1-2. revision='170746c' : Basic ChatVector, with 80B+ trained KoEn ckpt(rev. b4c45ab) .

v2. revision='4f04d1e' : Model diff based merging(ref. https://huggingface.co/blog/maywell/llm-feature-transfer ), with 25B+ trained KoEn ckpt(rev. d4d25a2) .

import requests
from PIL import Image

import torch
from transformers import AutoProcessor, LlavaForConditionalGeneration

model_id = "beomi/Llama-3-KoEn-8B-xtuner-llava-preview"

model = LlavaForConditionalGeneration.from_pretrained(
    model_id, 
    torch_dtype='auto', 
    device_map='auto',
    revision='a38aac3', # 'a38aac3' for basic ChatVector, '4f04d1e' for Model diff based merging(ref. https://huggingface.co/blog/maywell/llm-feature-transfer)
)

processor = AutoProcessor.from_pretrained(model_id)

tokenizer = processor.tokenizer
terminators = [
    tokenizer.eos_token_id,
    tokenizer.convert_tokens_to_ids("<|eot_id|>")
]

prompt = ("<|start_header_id|>user<|end_header_id|>\n\n<image>\n์ด ์ด๋ฏธ์ง€์— ๋Œ€ํ•ด์„œ ์„ค๋ช…ํ•ด์ฃผ์„ธ์š”.<|eot_id|>"
          "<|start_header_id|>assistant<|end_header_id|>\n\n์ด ์ด๋ฏธ์ง€์—๋Š”")
image_file = "https://cdn-uploads.huggingface.co/production/uploads/5e56829137cb5b49818287ea/NWfoArWI4UPAxpEnolkwT.jpeg"

raw_image = Image.open(requests.get(image_file, stream=True).raw)
inputs = processor(prompt, raw_image, return_tensors='pt').to(0, torch.float16)

output = model.generate(**inputs, max_new_tokens=400, do_sample=True, eos_token_id=terminators,)
print(processor.decode(output[0][2:], skip_special_tokens=False))

# --- Example Output [v1, Chat Vector] ---
user<|end_header_id|>

<image>
์ด ์ด๋ฏธ์ง€์— ๋Œ€ํ•ด์„œ ์„ค๋ช…ํ•ด์ฃผ์„ธ์š”.<|eot_id|><|start_header_id|>assistant<|end_header_id|>

์ด ์ด๋ฏธ์ง€์—๋Š” ๊ณ ์–‘์ด ํ•œ ๋งˆ๋ฆฌ๊ฐ€ ๊ฐ•๋ฌผ ์œ„๋ฅผ ๊ฑธ์–ด๊ฐ€๋Š” ๋ชจ์Šต์ด ๋ณด์—ฌ์ง‘๋‹ˆ๋‹ค. ๊ณ ์–‘์ด๋Š” ๊ฐ•๋ฌผ์˜ ์ž”๋ฌผ๊ฒฐ์— ๋ฏธ๋„๋Ÿผ์„ ํƒ€๊ณ  ๊ฐ• ๊ฐ€๋กœ๋ฅผ ์ง€๋‚˜๋Š” ๋ฐ ๋Šฅ์ˆ™ํ•˜๊ฒŒ ๋ณด์ž…๋‹ˆ๋‹ค. ๊ณ ์–‘์ด์˜ ๋ฐœ์€ ๊ฐ•๋ฌผ๋กœ ์ž˜ ๋“ค์–ด๊ฐ€, ๊ทธ๊ฒƒ์„ ์ฆ๊ธฐ๋ฉฐ ๊ฑธ์–ด๊ฐ‘๋‹ˆ๋‹ค. 

๋˜ํ•œ ์ด ์ด๋ฏธ์ง€๋„ ์Œ์„ฑ ๋…น์Œ์„ ํ•˜๊ฑฐ๋‚˜ ๋…นํ™”๋œ ์ž๋ฃŒ๋กœ ์ œ์ž‘๋˜์—ˆ์œผ๋ฉฐ, ์ฃผ๋กœ ๊ณ ์–‘์ด์˜ ๋ชจ์Šต์„ ๊ฐ•ํ•˜๊ฒŒ ๋ณด์—ฌ์ค๋‹ˆ๋‹ค. ์†Œ๋ฆฌ ํšจ๊ณผ๋„ ์—ฌ๋Ÿฌ ๊ฐ€์ง€๋กœ ์ถ”๊ฐ€ํ•˜์—ฌ ๊ณ ์–‘์ด์˜ ์Šคํ† ๋ฆฌ๋ฅผ ๋‹ค์–‘ํ•˜๊ฒŒ ์ „๋‹ฌํ•ฉ๋‹ˆ๋‹ค. ๊ฐ•๋ฌผ์€ ์ž”๋ฌผ๊ฒฐ์„ ๋‚˜ํƒ€๋‚ด๋ฉฐ ๊ฐ•๋ฌผ ์œ„๋ฅผ ๊ฑท๋Š” ๊ณ ์–‘์ด์˜ ๋ชจ์Šต์„ ๋”์šฑ ๊ฐ•๋ ฌํ•˜๊ฒŒ ๊ฐ•์กฐํ•˜๊ธฐ ์œ„ํ•ด ์ž”๋ฌผ๊ฒฐ์„ ํ†ตํ•ด ๋” ๋””ํ…Œ์ผํ•œ ์žฅ๋ฉด์„ ๋ณด์—ฌ์ค๋‹ˆ๋‹ค.<|eot_id|>

# --- Example Output [v1-1, Chat Vector] ---
user<|end_header_id|>

<image>
์ด ์ด๋ฏธ์ง€์— ๋Œ€ํ•ด์„œ ์„ค๋ช…ํ•ด์ฃผ์„ธ์š”.<|eot_id|><|start_header_id|>assistant<|end_header_id|>

์ด ์ด๋ฏธ์ง€์—์„œ๋Š” ํ•œ ๊ณ ์–‘์ด๊ฐ€ ์„œํ•ด์•ˆ์— ์œ„์น˜ํ•œ ๋ฐ”๋‹ค๋ฅผ ๊ฑท๊ณ  ์žˆ๋Š” ๋ชจ์Šต์„ ๋ณผ ์ˆ˜ ์žˆ์Šต๋‹ˆ๋‹ค. ๊ณ ์–‘์ด๋Š” ํ•ด๋ณ€์—์„œ๋ถ€ํ„ฐ ๋ฐ”๋‹ค๋กœ ๊ฑธ์–ด๋“ค์–ด๊ฐ€๋Š” ์ค‘์ด๋ฉฐ, ์ฃผ๋ณ€์—๋Š” ์ž”์ž”ํ•œ ํŒŒ๋„๊ฐ€ ๋ฐ€๋ ค์˜ค๋Š” ๋ชจ์Šต์„ ๋ณด์—ฌ์ฃผ๊ณ  ์žˆ์Šต๋‹ˆ๋‹ค. ์ด ๊ณ ์–‘์ด๋Š” ํƒœ์–ด๋‚  ๋•Œ๋ถ€ํ„ฐ ๊ณ ์–‘์ด์™€ ๊ฐ•์•„์ง€์™€๋Š” ๋‹ค๋ฅด๊ฒŒ ๋ฐ”๋‹ค๋ฅผ ๊ฒฝํ—˜ํ•˜๊ณ , ์ ์‘ํ•ด๊ฐ€๊ณ  ์žˆ์Šต๋‹ˆ๋‹ค. ๊ณ ์–‘์ด๋Š” ๋ฐ”๋‹ค๋ฅผ ์ข‹์•„ํ•˜๊ณ , ์ด ํ™˜๊ฒฝ์—์„œ ํ–‰๋ณต์„ ๋А๋ผ๋Š” ๊ฒƒ ๊ฐ™์Šต๋‹ˆ๋‹ค. ์ด ๊ณ ์–‘์ด๋Š” ์ธ๊ฐ„์ด ์•„๋‹Œ ์ž์—ฐ์˜ ์ผ๋ถ€๋กœ์จ ์ด ํ™˜๊ฒฝ์—์„œ ์‚ด์•„๊ฐ€๊ณ  ์žˆ์Šต๋‹ˆ๋‹ค.<|eot_id|>

# --- Example Output [v1-2, Chat Vector] ---
# model.generate(**inputs, max_new_tokens=200, do_sample=True, top_p=0.7, eos_token_id=terminators,)
user<|end_header_id|>

<image>
์ด ์ด๋ฏธ์ง€์— ๋Œ€ํ•ด์„œ ์„ค๋ช…ํ•ด์ฃผ์„ธ์š”.<|eot_id|><|start_header_id|>assistant<|end_header_id|>

์ด ์ด๋ฏธ์ง€๋Š” ํ•œ ๊ณ ์–‘์ด๊ฐ€ ๋ฌผ ์œ„๋ฅผ ๊ฑท๊ณ  ์žˆ๋Š” ๋ชจ์Šต์„ ํฌ์ฐฉํ•œ ์‚ฌ์ง„์ž…๋‹ˆ๋‹ค. ๊ณ ์–‘์ด๋Š” ๋‘ ๋ฐœ๋กœ ๋ฌผ ์œ„๋ฅผ ๊ฑธ์–ด ๊ฐ€๊ณ  ์žˆ์Šต๋‹ˆ๋‹ค. ๊ณ ์–‘์ด๋Š” 4๊ฐœ์˜ ๋ฐœ ์ค‘ 2๊ฐœ์˜ ๋ฐœ์€ ๋ฌผ์— ๋น ์ง€์ง€ ์•Š๊ณ  2๊ฐœ์˜ ๋ฐœ์€ ๋ฌผ์— ๋น ์ ธ ์žˆ์Šต๋‹ˆ๋‹ค. ๊ณ ์–‘์ด์˜ ๋ฐœ์ด ๋น ์ง„ ๋ถ€๋ถ„์€ ๋ฐ˜์˜๋˜์–ด ๋ฌผ์— ๋น„์ณ ์žˆ์Šต๋‹ˆ๋‹ค. ๋ฌผ ์œ„๋ฅผ ๊ฑท๋Š” ๊ณ ์–‘์ด์˜ ๋ชจ์Šต์ด ์ฐธ์œผ๋กœ ๊ท€์—ฝ๊ณ  ์‚ฌ๋ž‘์Šค๋Ÿฝ์Šต๋‹ˆ๋‹ค. ์ด ์‚ฌ์ง„์€ KBS ๋™๋ฌผ์˜ ์™•๊ตญ์—์„œ ๋ฐฉ์˜๋˜์—ˆ์Šต๋‹ˆ๋‹ค. KBS ๋™๋ฌผ์˜ ์™•๊ตญ์€ 1985๋…„๋ถ€ํ„ฐ ์‹œ์ž‘ํ•˜์—ฌ 2019๋…„๊นŒ์ง€ 34๋…„ ๋™์•ˆ ๋ฐฉ์˜๋œ KBS์˜ ๋Œ€ํ‘œ์ ์ธ ์ž์—ฐ ๋‹คํ๋ฉ˜ํ„ฐ๋ฆฌ ํ”„๋กœ๊ทธ๋žจ์ž…๋‹ˆ๋‹ค. KBS ๋™๋ฌผ์˜ ์™•๊ตญ์€ ๋™๋ฌผ์˜ ์ƒํƒœ์™€ ์Šต์„ฑ, ํ–‰๋™, ๊ทธ๋ฆฌ๊ณ  ์ž์—ฐ ํ™˜๊ฒฝ์„ ์ดํ•ดํ•˜๊ณ  ๋ณดํ˜ธํ•˜๋Š” ๋ฐ ๊ธฐ์—ฌํ•˜๊ณ ์ž ํ•ฉ๋‹ˆ๋‹ค.

# --- Example Output [v2, Model diff based merging] ---
user<|end_header_id|>

<image>
์ด ์ด๋ฏธ์ง€์— ๋Œ€ํ•ด์„œ ์„ค๋ช…ํ•ด์ฃผ์„ธ์š”.<|eot_id|><|start_header_id|>assistant<|end_header_id|>

์ด ์ด๋ฏธ์ง€์—๋Š” ํ•œ๊ตญ์–ด ์ž๋ง‰๊ณผ ํ•จ๊ป˜ ๊ณ ์–‘์ด๊ฐ€ ๋ฌผ์— ๋ฐœ์„ ๋””๋””๊ณ  ๊ฑท๋Š” ๋ชจ์Šต์ด ๋‹ด๊ฒจ ์žˆ์Šต๋‹ˆ๋‹ค. ๊ณ ์–‘์ด๋Š” ์˜ค๋ฅธ์ชฝ ๋ฐœ์„ ๋ฌผ์— ๋‹ด๊ทธ๊ณ  ๊ฑท๋Š” ์ค‘์ด๋ฉฐ, ํ•œ๊ตญ์–ด ์ž๋ง‰์€ "๊ณ ์–‘์ด๋Š” ๋ฌผ์„ ์ข‹์•„ํ•ฉ๋‹ˆ๋‹ค"๋ผ๋Š” ๋ฌธ์žฅ์„ ํฌํ•จํ•˜๊ณ  ์žˆ์Šต๋‹ˆ๋‹ค. ์ด ์ž๋ง‰์€ ๊ณ ์–‘์ด๊ฐ€ ๋ฌผ์„ ์ข‹์•„ํ•˜๋Š” ๊ฒƒ์„ ๊ฐ•์กฐํ•˜๊ณ  ์žˆ์Šต๋‹ˆ๋‹ค.<|eot_id|>

Runs of beomi Llama-3-KoEn-8B-xtuner-llava-preview on huggingface.co

86
Total runs
4
24-hour runs
4
3-day runs
12
7-day runs
67
30-day runs

More Information About Llama-3-KoEn-8B-xtuner-llava-preview huggingface.co Model

More Llama-3-KoEn-8B-xtuner-llava-preview license Visit here:

https://choosealicense.com/licenses/cc-by-nc-sa-4.0

Llama-3-KoEn-8B-xtuner-llava-preview huggingface.co

Llama-3-KoEn-8B-xtuner-llava-preview huggingface.co is an AI model on huggingface.co that provides Llama-3-KoEn-8B-xtuner-llava-preview's model effect (), which can be used instantly with this beomi Llama-3-KoEn-8B-xtuner-llava-preview model. huggingface.co supports a free trial of the Llama-3-KoEn-8B-xtuner-llava-preview model, and also provides paid use of the Llama-3-KoEn-8B-xtuner-llava-preview. Support call Llama-3-KoEn-8B-xtuner-llava-preview model through api, including Node.js, Python, http.

Llama-3-KoEn-8B-xtuner-llava-preview huggingface.co Url

https://huggingface.co/beomi/Llama-3-KoEn-8B-xtuner-llava-preview

beomi Llama-3-KoEn-8B-xtuner-llava-preview online free

Llama-3-KoEn-8B-xtuner-llava-preview huggingface.co is an online trial and call api platform, which integrates Llama-3-KoEn-8B-xtuner-llava-preview's modeling effects, including api services, and provides a free online trial of Llama-3-KoEn-8B-xtuner-llava-preview, you can try Llama-3-KoEn-8B-xtuner-llava-preview online for free by clicking the link below.

beomi Llama-3-KoEn-8B-xtuner-llava-preview online free url in huggingface.co:

https://huggingface.co/beomi/Llama-3-KoEn-8B-xtuner-llava-preview

Llama-3-KoEn-8B-xtuner-llava-preview install

Llama-3-KoEn-8B-xtuner-llava-preview is an open source model from GitHub that offers a free installation service, and any user can find Llama-3-KoEn-8B-xtuner-llava-preview on GitHub to install. At the same time, huggingface.co provides the effect of Llama-3-KoEn-8B-xtuner-llava-preview install, users can directly use Llama-3-KoEn-8B-xtuner-llava-preview installed effect in huggingface.co for debugging and trial. It also supports api for free installation.

Llama-3-KoEn-8B-xtuner-llava-preview install url in huggingface.co:

https://huggingface.co/beomi/Llama-3-KoEn-8B-xtuner-llava-preview

Url of Llama-3-KoEn-8B-xtuner-llava-preview

Llama-3-KoEn-8B-xtuner-llava-preview huggingface.co Url

Provider of Llama-3-KoEn-8B-xtuner-llava-preview huggingface.co

beomi
ORGANIZATIONS

Other API from beomi

huggingface.co

Total runs: 4.6K
Run Growth: 1.0K
Growth Rate: 21.64%
Updated:March 30 2023
huggingface.co

Total runs: 715
Run Growth: 464
Growth Rate: 64.90%
Updated:March 30 2023
huggingface.co

Total runs: 671
Run Growth: 344
Growth Rate: 51.27%
Updated:July 08 2024
huggingface.co

Total runs: 651
Run Growth: 190
Growth Rate: 29.19%
Updated:December 27 2023
huggingface.co

Total runs: 295
Run Growth: 165
Growth Rate: 55.93%
Updated:July 08 2024
huggingface.co

Total runs: 281
Run Growth: 264
Growth Rate: 93.95%
Updated:November 23 2021
huggingface.co

Total runs: 261
Run Growth: 136
Growth Rate: 52.11%
Updated:July 20 2023
huggingface.co

Total runs: 250
Run Growth: 167
Growth Rate: 66.80%
Updated:June 28 2023
huggingface.co

Total runs: 131
Run Growth: 21
Growth Rate: 15.56%
Updated:March 26 2024
huggingface.co

Total runs: 93
Run Growth: 75
Growth Rate: 80.65%
Updated:November 13 2023
huggingface.co

Total runs: 77
Run Growth: -139
Growth Rate: -182.89%
Updated:March 26 2024
huggingface.co

Total runs: 70
Run Growth: -17
Growth Rate: -24.29%
Updated:July 11 2023
huggingface.co

Total runs: 39
Run Growth: 26
Growth Rate: 66.67%
Updated:September 15 2023
huggingface.co

Total runs: 33
Run Growth: 4
Growth Rate: 12.12%
Updated:July 08 2024
huggingface.co

Total runs: 11
Run Growth: -19
Growth Rate: -172.73%
Updated:March 01 2022
huggingface.co

Total runs: 11
Run Growth: 0
Growth Rate: 0.00%
Updated:May 21 2021
huggingface.co

Total runs: 10
Run Growth: 2
Growth Rate: 20.00%
Updated:May 07 2023
huggingface.co

Total runs: 10
Run Growth: 3
Growth Rate: 30.00%
Updated:June 29 2023
huggingface.co

Total runs: 8
Run Growth: 0
Growth Rate: 0.00%
Updated:March 08 2023
huggingface.co

Total runs: 7
Run Growth: -2
Growth Rate: -28.57%
Updated:June 11 2021
huggingface.co

Total runs: 1
Run Growth: 1
Growth Rate: 100.00%
Updated:November 03 2023
huggingface.co

Total runs: 0
Run Growth: 0
Growth Rate: 0.00%
Updated:February 10 2022
huggingface.co

Total runs: 0
Run Growth: 0
Growth Rate: 0.00%
Updated:June 10 2021