Image-Text-to-Text
Transformers
Safetensors
Korean
English
gemma4
awaxis
korean
kr
gemma
gemma-4
Mixture of Experts
mixture-of-experts
multimodal
vision-language
darwin-derived
vidraft
darwin-crossbreed
agent
conversational
Eval Results (legacy)
Instructions to use Anserwise/AWAXIS-KR-31B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use Anserwise/AWAXIS-KR-31B with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("image-text-to-text", model="Anserwise/AWAXIS-KR-31B") messages = [ { "role": "user", "content": [ {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"}, {"type": "text", "text": "What animal is on the candy?"} ] }, ] pipe(text=messages)# pip install -U transformers accelerate # Load model directly from transformers import AutoProcessor, AutoModelForMultimodalLM processor = AutoProcessor.from_pretrained("Anserwise/AWAXIS-KR-31B") model = AutoModelForMultimodalLM.from_pretrained("Anserwise/AWAXIS-KR-31B", device_map="auto") messages = [ { "role": "user", "content": [ {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"}, {"type": "text", "text": "What animal is on the candy?"} ] }, ] inputs = processor.apply_chat_template( messages, add_generation_prompt=True, tokenize=True, return_dict=True, return_tensors="pt", ).to(model.device) outputs = model.generate(**inputs, max_new_tokens=256) print(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:])) - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use Anserwise/AWAXIS-KR-31B with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "Anserwise/AWAXIS-KR-31B" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "Anserwise/AWAXIS-KR-31B", "messages": [ { "role": "user", "content": [ { "type": "text", "text": "Describe this image in one sentence." }, { "type": "image_url", "image_url": { "url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg" } } ] } ] }'Use Docker
docker model run hf.co/Anserwise/AWAXIS-KR-31B
- SGLang
How to use Anserwise/AWAXIS-KR-31B with SGLang:
Install from pip and serve model
# Install SGLang from pip: pip install sglang # Start the SGLang server: python3 -m sglang.launch_server \ --model-path "Anserwise/AWAXIS-KR-31B" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "Anserwise/AWAXIS-KR-31B", "messages": [ { "role": "user", "content": [ { "type": "text", "text": "Describe this image in one sentence." }, { "type": "image_url", "image_url": { "url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg" } } ] } ] }'Use Docker images
docker run --gpus all \ --shm-size 32g \ -p 30000:30000 \ -v ~/.cache/huggingface:/root/.cache/huggingface \ --env "HF_TOKEN=<secret>" \ --ipc=host \ lmsysorg/sglang:latest \ python3 -m sglang.launch_server \ --model-path "Anserwise/AWAXIS-KR-31B" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "Anserwise/AWAXIS-KR-31B", "messages": [ { "role": "user", "content": [ { "type": "text", "text": "Describe this image in one sentence." }, { "type": "image_url", "image_url": { "url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg" } } ] } ] }' - Docker Model Runner
How to use Anserwise/AWAXIS-KR-31B with Docker Model Runner:
docker model run hf.co/Anserwise/AWAXIS-KR-31B
docs: Add VIDRAFT Darwin platform breeding/evolution description
Browse files
README.md
CHANGED
|
@@ -18,6 +18,8 @@ tags:
|
|
| 18 |
- vision-language
|
| 19 |
- image-text-to-text
|
| 20 |
- darwin-derived
|
|
|
|
|
|
|
| 21 |
- agent
|
| 22 |
base_model:
|
| 23 |
- google/gemma-4-31B-it
|
|
@@ -55,72 +57,116 @@ model-index:
|
|
| 55 |
|
| 56 |
# AWAXIS-KR-31B
|
| 57 |
|
| 58 |
-
##
|
| 59 |
|
| 60 |
-
**AWAXIS-KR-31B**은 한국어 특화 MoE 베이스(JDONE-Research/AIOne-Agent-52B-A36B-it)에 Opus-distill 추론 시그널(Anserwise/AWAXIS-Think-31B)을 결합
|
| 61 |
|
| 62 |
-
|
|
|
|
|
|
|
| 63 |
|
| 64 |
---
|
| 65 |
|
| 66 |
-
##
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 67 |
|
| 68 |
```
|
| 69 |
-
AWAXIS-
|
| 70 |
-
|
| 71 |
-
|
| 72 |
-
|
| 73 |
-
|
| 74 |
-
|
| 75 |
-
|
| 76 |
-
|
| 77 |
-
|
| 78 |
-
│ — Claude Opus 추론 distill 베이스
|
| 79 |
-
│
|
| 80 |
-
└── 조부 (FFN donor)
|
| 81 |
-
└── google/gemma-4-31B-it
|
| 82 |
-
— Gemma-4 베이스
|
| 83 |
```
|
| 84 |
|
| 85 |
-
|
| 86 |
|
| 87 |
-
|
| 88 |
-
|------|------|------|
|
| 89 |
-
| 어머니 Mother (kept) | [JDONE-Research/AIOne-Agent-52B-A36B-it](https://huggingface.co/JDONE-Research/AIOne-Agent-52B-A36B-it) | 한국어 능력, MoE 라우팅, 전문가, 어텐션, 임베딩 100% 보존 |
|
| 90 |
-
| 아버지 Father (FFN donor) | [Anserwise/AWAXIS-Think-31B](https://huggingface.co/Anserwise/AWAXIS-Think-31B) | Opus-distill 추론 시그널을 dense FFN 경로로 주입 |
|
| 91 |
|
| 92 |
-
|
| 93 |
|
| 94 |
-
|
| 95 |
-
|
| 96 |
-
|
| 97 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 98 |
|
| 99 |
-
**
|
| 100 |
|
| 101 |
---
|
| 102 |
|
| 103 |
-
##
|
| 104 |
|
| 105 |
본 모델의 **한국어 능력 평가**에는 **K-AI Hub(NIA AI Hub) / K-AI Leaderboard(aihub.or.kr) 생태계**의 표준 한국어 LLM 벤치마크 데이터셋을 활용했습니다.
|
| 106 |
|
| 107 |
-
|
|
| 108 |
-
|---|---|---|
|
| 109 |
-
| **KMMLU** |
|
| 110 |
-
| **HAE_RAE_BENCH_1.1** |
|
| 111 |
-
| **HRM8K** |
|
| 112 |
-
| **CLIcK** |
|
| 113 |
-
|
| 114 |
-
상기 데이터셋은 HAERAE-HUB와 EunsuKim 등 한국 연구 커뮤니티가 큐레이팅하여 K-AI 허브 평가 표준으로 채택된 공공 자산입니다.
|
| 115 |
|
| 116 |
---
|
| 117 |
|
| 118 |
-
##
|
| 119 |
|
| 120 |
| | |
|
| 121 |
|---|---|
|
| 122 |
| Class | `Gemma4ForConditionalGeneration` (multimodal: text + image + audio) |
|
| 123 |
-
| Parameters | **52B total
|
| 124 |
| Layers | 60 |
|
| 125 |
| Hidden / Intermediate | 5,376 / 21,504 |
|
| 126 |
| Attention heads / head_dim | 32 / 256 |
|
|
@@ -129,29 +175,29 @@ AWAXIS-KR-31B (this model — Darwin-derived)
|
|
| 129 |
|
| 130 |
---
|
| 131 |
|
| 132 |
-
##
|
| 133 |
|
| 134 |
-
|
|
| 135 |
|-----------|---------|-------|
|
| 136 |
-
| **
|
| 137 |
-
|
|
| 138 |
-
|
|
| 139 |
-
|
|
| 140 |
-
|
|
| 141 |
| **CLIcK** (n=200) | greedy | **88.0%** |
|
| 142 |
|
| 143 |
---
|
| 144 |
|
| 145 |
-
##
|
| 146 |
|
| 147 |
-
-
|
| 148 |
-
-
|
| 149 |
-
-
|
| 150 |
-
-
|
| 151 |
|
| 152 |
---
|
| 153 |
|
| 154 |
-
##
|
| 155 |
|
| 156 |
```python
|
| 157 |
from transformers import AutoTokenizer, AutoModelForCausalLM
|
|
@@ -175,6 +221,17 @@ print(tok.decode(out[0][inp["input_ids"].shape[-1]:], skip_special_tokens=True))
|
|
| 175 |
|
| 176 |
---
|
| 177 |
|
| 178 |
-
##
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 179 |
|
| 180 |
-
|
|
|
|
| 18 |
- vision-language
|
| 19 |
- image-text-to-text
|
| 20 |
- darwin-derived
|
| 21 |
+
- vidraft
|
| 22 |
+
- darwin-crossbreed
|
| 23 |
- agent
|
| 24 |
base_model:
|
| 25 |
- google/gemma-4-31B-it
|
|
|
|
| 57 |
|
| 58 |
# AWAXIS-KR-31B
|
| 59 |
|
| 60 |
+
## Overview
|
| 61 |
|
| 62 |
+
**AWAXIS-KR-31B**은 **[VIDRAFT](https://huggingface.co/VIDraft) Darwin AI 모델 교배/진화 플랫폼**을 통해 생성된 한국어 특화 MoE 모델입니다. Darwin의 독자적인 **FFN-crossbreed 엔진(V8)**으로 한국어 특화 MoE 베이스(JDONE-Research/AIOne-Agent-52B-A36B-it)에 Opus-distill 추론 시그널(Anserwise/AWAXIS-Think-31B)을 교배 결합하였습니다.
|
| 63 |
|
| 64 |
+
Gemma-4 MoE 아키텍처(8 전문가 top-2 라우팅, **52B 총 / 36B 활성** 파라미터, vision/audio 토큰 지원) 기반으로, 한국어 instruction following, 지식/문화 QA, 단계별 추론/수학 작업에 최적화되어 있으며, 한국어 4과목 종합 **80.0%** 성능을 검증했습니다.
|
| 65 |
+
|
| 66 |
+
> AWAXIS-KR-31B is a Korean-focused MoE model (Gemma-4 family, 52B total / 36B active, 8 experts top-2 routing) created through the **[VIDRAFT](https://huggingface.co/VIDraft) Darwin AI Model Breeding/Evolution Platform**. Built via Darwin V8 FFN-crossbreed engine, combining a Korean-specialized MoE base with Opus-distill reasoning signals through automated biological-inspired crossbreeding.
|
| 67 |
|
| 68 |
---
|
| 69 |
|
| 70 |
+
## VIDRAFT Darwin AI 모델 교배/진화 플랫폼
|
| 71 |
+
|
| 72 |
+
**[VIDRAFT Darwin](https://huggingface.co/VIDraft)**은 AI 모델의 **교배(Crossbreeding)와 진화(Evolution)**를 통해 새로운 고성능 모델을 자동 생성하는 플랫폼입니다. 생물학적 유전 원리에서 영감을 받아, 두 개 이상의 부모 모델에서 각각의 장점을 선택적으로 결합하여 자식 모델을 탄생시킵니다.
|
| 73 |
+
|
| 74 |
+
### Darwin 교배/진화 핵심 기술
|
| 75 |
+
|
| 76 |
+
| 기술 | 설명 |
|
| 77 |
+
|------|------|
|
| 78 |
+
| **FFN Crossbreed Engine (V8)** | 부모 모델의 Feed-Forward Network(FFN) 레이어를 선택적으로 교차 결합하는 핵심 엔진. 어텐션/임베딩은 어머니(Mother)에서, FFN 시그널은 아버지(Father)에서 추출하여 블렌딩 |
|
| 79 |
+
| **Smart MRI (Model Resonance Imaging)** | 두 모델 간 레이어별 유사도/호환성을 분석하여 최적 교배 비율(alpha)을 자동 탐색하는 기술 |
|
| 80 |
+
| **Alpha Grid Search** | 교배 비율 alpha를 체계적으로 탐색하여 벤치마크 성능이 최대화되는 최적점을 발견 (자연선택 시뮬레이션) |
|
| 81 |
+
| **Multi-Generation Breeding** | 1세대 교배 결과물을 다시 부모로 삼아 2세대, 3세대 교배를 수행하는 다세대 진화 |
|
| 82 |
+
|
| 83 |
+
### 이 모델의 Darwin 교배 과정
|
| 84 |
+
|
| 85 |
+
**AWAXIS-KR-31B은 2세대(F2) 교배 모델**입니다. 1세대에서 AWAXIS-Think-31B을 생성하고, 이를 다시 아버지로 삼아 한국어 MoE 어머니와 2세대 교배를 수행했습니다.
|
| 86 |
|
| 87 |
```
|
| 88 |
+
[1세대 교배] AWAXIS-Think-31B 생성
|
| 89 |
+
Mother: TeichAI/gemma-4-31B-it-Claude-Opus-Distill-v2
|
| 90 |
+
Father: google/gemma-4-31B-it
|
| 91 |
+
--> Darwin FFN-crossbreed (alpha=0.1) --> AWAXIS-Think-31B
|
| 92 |
+
|
| 93 |
+
[2세대 교배] AWAXIS-KR-31B 생성 (이 모델)
|
| 94 |
+
Mother: JDONE-Research/AIOne-Agent-52B-A36B-it (한국어 MoE)
|
| 95 |
+
Father: AWAXIS-Think-31B (1세대 교배 결과물)
|
| 96 |
+
--> Darwin FFN-crossbreed --> AWAXIS-KR-31B
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 97 |
```
|
| 98 |
|
| 99 |
+
이처럼 Darwin 플랫폼은 **세대를 거듭할수록 능력이 누적 진화**하는 다세대 교배(Multi-Generation Breeding)를 지원합니다.
|
| 100 |
|
| 101 |
+
### 왜 Darwin 교배인가?
|
|
|
|
|
|
|
|
|
|
| 102 |
|
| 103 |
+
기존 모델 합성 방식(단순 가중치 평균, SLERP, TIES 등)과 달리, Darwin 교배는:
|
| 104 |
|
| 105 |
+
1. **생물학적 유전 모방**: 어머니/아버지 역할을 명확히 분리하여 각 부모의 핵심 능력만 선택적으로 상속
|
| 106 |
+
2. **FFN 선택적 주입**: 어텐션(문맥 이해)은 어머니에서 100% 보존하고, FFN(지식/추론 패턴)만 아버지에서 교차 -> 능력 충돌 최소화
|
| 107 |
+
3. **벤치마크 기반 자연선택**: alpha grid search로 여러 자식 후보를 생성한 뒤, 실측 벤치마크로 최적 개체를 선택
|
| 108 |
+
4. **다세대 진화**: 1세대 결과를 부모로 재활용하여 능력 누적 (이 모델 = 2세대)
|
| 109 |
+
|
| 110 |
+
---
|
| 111 |
+
|
| 112 |
+
## Model Lineage (모델 족보)
|
| 113 |
+
|
| 114 |
+
```
|
| 115 |
+
AWAXIS-KR-31B (this model -- 2nd generation Darwin crossbreed)
|
| 116 |
+
|
|
| 117 |
+
+-- Mother (kept full, 100%)
|
| 118 |
+
| JDONE-Research/AIOne-Agent-52B-A36B-it
|
| 119 |
+
| -- Korean-specialized Gemma4 MoE 52B / A36B
|
| 120 |
+
|
|
| 121 |
+
+-- Father (FFN donor)
|
| 122 |
+
Anserwise/AWAXIS-Think-31B (1st generation Darwin crossbreed)
|
| 123 |
+
|
|
| 124 |
+
+-- Grandmother (kept full)
|
| 125 |
+
| TeichAI/gemma-4-31B-it-Claude-Opus-Distill-v2
|
| 126 |
+
| -- Claude Opus reasoning distill base
|
| 127 |
+
|
|
| 128 |
+
+-- Grandfather (FFN donor)
|
| 129 |
+
google/gemma-4-31B-it
|
| 130 |
+
-- Gemma-4 base
|
| 131 |
+
```
|
| 132 |
+
|
| 133 |
+
### Direct Parents
|
| 134 |
+
|
| 135 |
+
| Role | Model | Contribution |
|
| 136 |
+
|------|-------|-------------|
|
| 137 |
+
| Mother (kept) | [JDONE-Research/AIOne-Agent-52B-A36B-it](https://huggingface.co/JDONE-Research/AIOne-Agent-52B-A36B-it) | Korean capability, MoE routing, experts, attention, embeddings 100% preserved |
|
| 138 |
+
| Father (FFN donor) | [Anserwise/AWAXIS-Think-31B](https://huggingface.co/Anserwise/AWAXIS-Think-31B) | Opus-distill reasoning signal injected via dense FFN pathway |
|
| 139 |
+
|
| 140 |
+
### Paternal Grandparents
|
| 141 |
+
|
| 142 |
+
| Role | Model |
|
| 143 |
+
|------|-------|
|
| 144 |
+
| Grandmother | [TeichAI/gemma-4-31B-it-Claude-Opus-Distill-v2](https://huggingface.co/TeichAI/gemma-4-31B-it-Claude-Opus-Distill-v2) |
|
| 145 |
+
| Grandfather | [google/gemma-4-31B-it](https://huggingface.co/google/gemma-4-31B-it) |
|
| 146 |
|
| 147 |
+
**Common ancestor**: Google **Gemma-4** architecture.
|
| 148 |
|
| 149 |
---
|
| 150 |
|
| 151 |
+
## Datasets Used (활용 데이터셋)
|
| 152 |
|
| 153 |
본 모델의 **한국어 능력 평가**에는 **K-AI Hub(NIA AI Hub) / K-AI Leaderboard(aihub.or.kr) 생태계**의 표준 한국어 LLM 벤치마크 데이터셋을 활용했습니다.
|
| 154 |
|
| 155 |
+
| Dataset | Domain | Source |
|
| 156 |
+
|---------|--------|--------|
|
| 157 |
+
| **KMMLU** | Korean knowledge (45 subjects) | [HAERAE-HUB/KMMLU](https://huggingface.co/datasets/HAERAE-HUB/KMMLU) |
|
| 158 |
+
| **HAE_RAE_BENCH_1.1** | Korean comprehension/culture (13 subsets) | [HAERAE-HUB/HAE_RAE_BENCH_1.1](https://huggingface.co/datasets/HAERAE-HUB/HAE_RAE_BENCH_1.1) |
|
| 159 |
+
| **HRM8K** | Korean math/reasoning (GSM8K Korean) | [HAERAE-HUB/HRM8K](https://huggingface.co/datasets/HAERAE-HUB/HRM8K) |
|
| 160 |
+
| **CLIcK** | Korean culture-language | [EunsuKim/CLIcK](https://huggingface.co/datasets/EunsuKim/CLIcK) |
|
|
|
|
|
|
|
| 161 |
|
| 162 |
---
|
| 163 |
|
| 164 |
+
## Architecture
|
| 165 |
|
| 166 |
| | |
|
| 167 |
|---|---|
|
| 168 |
| Class | `Gemma4ForConditionalGeneration` (multimodal: text + image + audio) |
|
| 169 |
+
| Parameters | **52B total / 36B active** (MoE, 8 experts, top-2 routing) |
|
| 170 |
| Layers | 60 |
|
| 171 |
| Hidden / Intermediate | 5,376 / 21,504 |
|
| 172 |
| Attention heads / head_dim | 32 / 256 |
|
|
|
|
| 175 |
|
| 176 |
---
|
| 177 |
|
| 178 |
+
## Measured Benchmarks
|
| 179 |
|
| 180 |
+
| Benchmark | Setting | Score |
|
| 181 |
|-----------|---------|-------|
|
| 182 |
+
| **Korean 4-Subject Composite** (n=80, seed=42) | greedy | **80.0%** |
|
| 183 |
+
| -- KMMLU (knowledge) | 20Q, greedy | 70.0% |
|
| 184 |
+
| -- HAERAE-Bench (comprehension) | 20Q, greedy | 75.0% |
|
| 185 |
+
| -- HRM8K (math) | 20Q, greedy | **90.0%** |
|
| 186 |
+
| -- CLIcK (culture-language) | 20Q, greedy | 85.0% |
|
| 187 |
| **CLIcK** (n=200) | greedy | **88.0%** |
|
| 188 |
|
| 189 |
---
|
| 190 |
|
| 191 |
+
## Intended Use
|
| 192 |
|
| 193 |
+
- Korean instruction following
|
| 194 |
+
- Knowledge/culture QA, reasoning/math
|
| 195 |
+
- General Korean LLM tasks
|
| 196 |
+
- Multimodal input (image-text-to-text) inherited from Gemma-4 base capability
|
| 197 |
|
| 198 |
---
|
| 199 |
|
| 200 |
+
## Inference
|
| 201 |
|
| 202 |
```python
|
| 203 |
from transformers import AutoTokenizer, AutoModelForCausalLM
|
|
|
|
| 221 |
|
| 222 |
---
|
| 223 |
|
| 224 |
+
## License
|
| 225 |
+
|
| 226 |
+
This model includes **Gemma-4 lineage** weights and complies with the [Gemma Terms of Use](https://ai.google.dev/gemma/terms).
|
| 227 |
+
|
| 228 |
+
## Acknowledgements
|
| 229 |
+
|
| 230 |
+
- **[VIDRAFT](https://huggingface.co/VIDraft)** -- Darwin AI Model Breeding/Evolution Platform
|
| 231 |
+
- JDONE-Research for the Korean MoE base
|
| 232 |
+
- TeichAI for the Opus-Distill base
|
| 233 |
+
- Google DeepMind for Gemma-4
|
| 234 |
+
|
| 235 |
+
---
|
| 236 |
|
| 237 |
+
*Built with the **VIDRAFT Darwin AI Model Breeding/Evolution Platform** -- FFN-crossbreed V8 engine. This is a 2nd-generation (F2) Darwin crossbreed model, created through automated biological-inspired crossbreeding that selectively combines the strengths of parent models. The Father (AWAXIS-Think-31B) was itself a 1st-generation Darwin crossbreed, demonstrating multi-generation evolution capability. Measured numbers above are exact; nothing inflated.*
|