Cogito V2 Preview Llama 109B
An instruction-tuned, hybrid-reasoning Mixture-of-Experts model built on Llama-4-Scout-17B-16E. Cogito v2 can answer directly or engage an extended “thinking” phase, with alignment guided by Iterated Distillation & Amplification (IDA). It targets coding, STEM, instruction following, and general helpfulness, with stronger multilingual, tool-calling, and reasoning performance than size-equivalent baselines. The model supports long-context use (up to 10M tokens) and standard Transformers workflows. Users can control the reasoning behaviour with the `reasoning` `enabled` boolean. [Learn more in our docs](https://openrouter.ai/docs/use-cases/reasoning-tokens#enable-reasoning-with-default-config)
OpenRouter 원가 (1M 토큰)
인보이스랩 (월 예상)
입력 0.5M + 출력 0.5M 기준
모델 정보
기본 정보
| 모델 ID | deepcogito/cogito-v2-preview-llama-109b-moe |
| 제공사 | Deepcogito |
| 컨텍스트 윈도우 | 32,767 토큰 |
| 모달리티 | text+image->text |
지원 기능
API 사용법
Python (OpenAI SDK 호환)
from openai import OpenAI
client = OpenAI(
api_key="your-dream-api-key",
base_url="https://api.invoicedream.co.kr/v1"
)
response = client.chat.completions.create(
model="deepcogito/cogito-v2-preview-llama-109b-moe",
messages=[
{"role": "user", "content": "안녕하세요"}
]
)
print(response.choices[0].message.content)Node.js / TypeScript
import OpenAI from 'openai';
const client = new OpenAI({
apiKey: 'your-dream-api-key',
baseURL: 'https://api.invoicedream.co.kr/v1'
});
const response = await client.chat.completions.create({
model: 'deepcogito/cogito-v2-preview-llama-109b-moe',
messages: [{ role: 'user', content: '안녕하세요' }]
});
console.log(response.choices[0].message.content);cURL
curl https://api.invoicedream.co.kr/v1/chat/completions \
-H "Authorization: Bearer your-dream-api-key" \
-H "Content-Type: application/json" \
-d '{
"model": "deepcogito/cogito-v2-preview-llama-109b-moe",
"messages": [{"role": "user", "content": "안녕하세요"}]
}'💡 Tip: OpenAI SDK를 그대로 사용할 수 있습니다.base_url만 변경하면 됩니다!